|
创作新主题 |
社区所有版块导航
DATA
| docker Elasticsearch |
WEB开发
| linux MongoDB Redis DATABASE NGINX 其他Web框架 web工具 zookeeper tornado NoSql Bootstrap js peewee Git bottle IE MQ Jquery |
机器学习
| 机器学习算法 |
产品
| 短视频 |
印度
| 印度 |
一周十大热门主题
点击阅读原文访问arxivdaily.com，涵盖CS|物理|数学|经济|统计|金融|生物|电气领域，更有搜索、收藏等功能！
cs.LG 方向，今日共计158篇
大模型相关(15篇)
【1】Do Natural Language Descriptions of Model Activations Convey Privileged Information?
标题：模型激活的自然语言描述是否传达了特权信息？
链接：https://arxiv.org/abs/2509.13316
备注：34 pages, 6 figures
摘要：最近的可解释性方法已经提出使用第二个动词化器LLM将LLM内部表示转换为自然语言描述。这旨在说明目标模型如何表示和操作输入。但是，这种激活语言化方法实际上提供了关于目标模型内部工作的特权知识，还是仅仅传达了关于其输入的信息？我们批判性地评估了在先前工作中使用的数据集上流行的语言化方法，发现它们在没有任何目标模型内部访问的情况下在基准测试中取得了成功，这表明这些数据集对于评估语言化方法并不理想。然后，我们运行控制实验，揭示了语言化往往反映了参数的语言化LLM生成它们的知识，而不是激活的目标LLM被解码。总之，我们的研究结果表明，需要有针对性的基准和实验控制，严格评估是否言语化方法提供有意义的见解LLM的操作。
摘要：Recent interpretability methods have proposed to translate LLM internal representations into natural language descriptions using a second verbalizer LLM. This is intended to illuminate how the target model represents and operates on inputs. But do such activation verbalization approaches actually provide privileged knowledge about the internal workings of the target model, or do they merely convey information about its inputs? We critically evaluate popular verbalization methods across datasets used in prior work and find that they succeed at benchmarks without any access to target model internals, suggesting that these datasets are not ideal for evaluating verbalization methods. We then run controlled experiments which reveal that verbalizations often reflect the parametric knowledge of the verbalizer LLM which generated them, rather than the activations of the target LLM being decoded. Taken together, our results indicate a need for targeted benchmarks and experimental controls to rigorously assess whether verbalization methods provide meaningful insights into the operations of LLMs.
【2】LLMs for energy and macronutrients estimation using only text data from 24-hour dietary recalls: a parameter-efficient fine-tuning experiment using a 10-shot prompt
标题：仅使用24小时饮食回忆的文本数据进行能量和大量营养素估算的LLM：使用10次提示的参数高效微调实验
链接：https://arxiv.org/abs/2509.13268
备注：his https URL
摘要：背景：大多数用于估计营养成分的人工智能工具都依赖于图像输入。然而，大型语言模型（LLM）是否可以仅仅基于所食用食物的文本描述来准确预测营养价值仍然是未知的。如果有效，这种方法可以实现更简单的饮食监测，而无需照片。方法：我们使用了国家健康与营养调查（NHANES）中12-19岁青少年的24小时饮食回忆。一个开源的量化LLM是使用10个镜头，链式思维方法来估计能量和五种宏量营养素，仅仅基于列出食物及其数量的文本字符串。然后，我们应用参数高效微调（PEFT）来评估预测准确性是否提高。NHANES计算的值作为能量、蛋白质、碳水化合物、总糖、膳食纤维和总脂肪的基本事实。结果：在11，281名青少年（49.9%为男性，平均年龄15.4岁）的汇总数据集中，香草LLM的预测效果较差。能量的平均绝对误差（MAE）为652.08，各终点的Lin's CCC <0.46。相比之下，微调模型的表现要好得多，能量MAE范围从171.34到190.90，Lin的CCC超过0.89。结论：当使用思想链方法提示并使用PEFT进行微调时，仅暴露于文本输入的开源LLM可以准确地预测24小时饮食回忆的能量和常量营养素值。这种方法有望成为低负担、基于文本的饮食监测工具。
摘要：BACKGROUND: Most artificial intelligence tools used to estimate nutritional content rely on image input. However, whether large language models (LLMs) can accurately predict nutritional values based solely on text descriptions of foods consumed remains unknown. If effective, this approach could enable simpler dietary monitoring without the need for photographs. METHODS: We used 24-hour dietary recalls from adolescents aged 12-19 years in the National Health and Nutrition Examination Survey (NHANES). An open-source quantized LLM was prompted using a 10-shot, chain-of-thought approach to estimate energy and five macronutrients based solely on text strings listing foods and their quantities. We then applied parameter-efficient fine-tuning (PEFT) to evaluate whether predictive accuracy improved. NHANES-calculated values served as the ground truth for energy, proteins, carbohydrates, total sugar, dietary fiber and total fat. RESULTS: In a pooled dataset of 11,281 adolescents (49.9% male, mean age 15.4 years), the vanilla LLM yielded poor predictions. The mean absolute error (MAE) was 652.08 for energy and the Lin's CCC <0.46 across endpoints. In contrast, the fine-tuned model performed substantially better, with energy MAEs ranging from 171.34 to 190.90 across subsets, and Lin's CCC exceeding 0.89 for all outcomes. CONCLUSIONS: When prompted using a chain-of-thought approach and fine-tuned with PEFT, open-source LLMs exposed solely to text input can accurately predict energy and macronutrient values from 24-hour dietary recalls. This approach holds promise for low-burden, text-based dietary monitoring tools.
【3】Metacognitive Reuse: Turning Recurring LLM Reasoning Into Concise Behaviors
标题：元认知重用：将反复出现的LLM推理转化为简洁的行为
链接：https://arxiv.org/abs/2509.13237
备注：18 pages, 9 Figures, 5 Tables
摘要：大型语言模型（LLM）现在通过发出扩展的思想链来解决多步问题。在这个过程中，他们经常在问题中重新推导相同的中间步骤，从而增加了令牌的使用和延迟。上下文窗口的这种饱和使得用于探索的容量更少。我们研究了一种简单的机制，通过模型自己的元语义分析先前的痕迹，将重复出现的推理片段转换为简洁，可重用的“行为”（名称+指令）。这些行为存储在“行为手册”中，该手册在推理时将它们提供给模型，或者通过监督微调将它们提取为参数。这种方法在三种不同的设置中实现了改进的测试时间推理- 1）行为条件推理：在推理期间在上下文中提供LLM相关行为将推理令牌的数量减少了高达46%，同时匹配或提高基线准确性; 2）行为指导的自我改进：在没有任何参数更新的情况下，模型通过利用自己过去解决问题尝试的行为来改进自己的未来推理。这产生了高达10%的准确性比一个天真的批评和修改基线;和3）行为条件SFT：行为条件推理痕迹的SFT是更有效的转换非推理模型到推理模型相比，香草SFT。总之，这些结果表明，将缓慢的推导转化为快速的程序提示，使LLM能够记住如何推理，而不仅仅是得出什么结论。
摘要 ：Large language models (LLMs) now solve multi-step problems by emitting extended chains of thought. During the process, they often re-derive the same intermediate steps across problems, inflating token usage and latency. This saturation of the context window leaves less capacity for exploration. We study a simple mechanism that converts recurring reasoning fragments into concise, reusable "behaviors" (name + instruction) via the model's own metacognitive analysis of prior traces. These behaviors are stored in a "behavior handbook" which supplies them to the model in-context at inference or distills them into parameters via supervised fine-tuning. This approach achieves improved test-time reasoning across three different settings - 1) Behavior-conditioned inference: Providing the LLM relevant behaviors in-context during reasoning reduces number of reasoning tokens by up to 46% while matching or improving baseline accuracy; 2) Behavior-guided self-improvement: Without any parameter updates, the model improves its own future reasoning by leveraging behaviors from its own past problem solving attempts. This yields up to 10% higher accuracy than a naive critique-and-revise baseline; and 3) Behavior-conditioned SFT: SFT on behavior-conditioned reasoning traces is more effective at converting non-reasoning models into reasoning models as compared to vanilla SFT. Together, these results indicate that turning slow derivations into fast procedural hints enables LLMs to remember how to reason, not just what to conclude.
【4】Efficient Cold-Start Recommendation via BPE Token-Level Embedding Initialization with LLM
标题：通过BPE代币级嵌入收件箱和LLM提供高效冷启动推荐
链接：https://arxiv.org/abs/2509.13179
摘要：当我们谈论推荐系统时，冷启动问题是一个挑战，特别是在我们没有新用户或新项目的过去交互数据的情况下。基于内容的功能或混合解决方案与传统解决方案一样常见，但它们只能在具有浅模式的稀疏元数据环境中工作。本文提出了一种高效的冷启动推荐策略，该策略基于子词级表示，在初始化过程中应用字节对编码（BPE）标记化和预训练的大语言模型（LLM）嵌入。我们获得细粒度的标记级向量，与BPE词汇表对齐，而不是使用粗粒度的句子嵌入。总之，这些令牌嵌入可以用作不可见实体的密集语义先验，从而在没有用户与项目交互历史的情况下实现即时推荐性能。我们的机制可以与协同过滤系统进行比较，并在具有严格冷启动假设的基准数据集上进行测试。实验结果表明，给定的BPE-LLM方法实现了更高的召回@k，NDCG@k，和命中率测量相比，标准基线，并显示足够的计算性能的相同的能力。此外，我们证明了使用子词感知嵌入产生更好的泛化能力，更可解释，特别是在多语言和稀疏输入设置。标记级语义初始化作为一个轻量级的，但仍然有效的扩展到现代推荐系统中的zero-shot设置的实际应用表明在这项工作。
摘要：The cold-start issue is the challenge when we talk about recommender systems, especially in the case when we do not have the past interaction data of new users or new items. Content-based features or hybrid solutions are common as conventional solutions, but they can only work in a sparse metadata environment with shallow patterns. In this paper, the efficient cold-start recommendation strategy is presented, which is based on the sub word-level representations by applying Byte Pair Encoding (BPE) tokenization and pre-trained Large Language Model (LLM) embedding in the initialization procedure. We obtain fine-grained token-level vectors that are aligned with the BPE vocabulary as opposed to using coarse-grained sentence embeddings. Together, these token embeddings can be used as dense semantic priors on unseen entities, making immediate recommendation performance possible without user-item interaction history. Our mechanism can be compared to collaborative filtering systems and tested over benchmark datasets with stringent cold-start assumptions. Experimental findings show that the given BPE-LLM method achieves higher Recall@k, NDCG@k, and Hit Rate measurements compared to the standard baseline and displays the same capability of sufficient computational performance. Furthermore, we demonstrate that using subword-aware embeddings yields better generalizability and is more interpretable, especially within a multilingual and sparse input setting. The practical application of token-level semantic initialization as a lightweight, but nevertheless effective extension to modern recommender systems in the zero-shot setting is indicated within this work.
【5】Discovering Mathematical Equations with Diffusion Language Model
标题：用扩散语言模型发现数学方程
链接：https://arxiv.org/abs/2509.13136
摘要：从观测数据中发现有效和有意义的数学方程在科学发现中起着至关重要的作用。虽然这个任务，符号回归，仍然具有挑战性，由于巨大的搜索空间和准确性和复杂性之间的权衡。在本文中，我们介绍了DiffuSR，这是一个建立在连续状态扩散语言模型上的符号回归预训练框架。DiffuSR在扩散过程中采用可训练的嵌入层，将离散的数学符号映射到连续的潜在空间，有效地建模方程分布。通过迭代去噪，DiffuSR将初始噪声序列转换为符号方程，并通过交叉注意机制注入数值数据。我们还设计了一个有效的推理策略，以提高基于扩散的方程发生器，注入logit先验遗传编程的准确性。在标准符号回归基准上的实验结果表明，DiffuSR与最先进的自回归方法相比具有竞争力的性能，并生成更可解释和更多样化的数学表达式。
摘要：Discovering valid and meaningful mathematical equations from observed data plays a crucial role in scientific discovery. While this task, symbolic regression, remains challenging due to the vast search space and the trade-off between accuracy and complexity. In this paper, we introduce DiffuSR, a pre-training framework for symbolic regression built upon a continuous-state diffusion language model. DiffuSR employs a trainable embedding layer within the diffusion process to map discrete mathematical symbols into a continuous latent space, modeling equation distributions effectively. Through iterative denoising, DiffuSR converts an initial noisy sequence into a symbolic equation, guided by numerical data injected via a cross-attention mechanism. We also design an effective inference strategy to enhance the accuracy of the diffusion-based equation generator, which injects logit priors into genetic programming. Experimental results on standard symbolic regression benchmarks demonstrate that DiffuSR achieves competitive performance with state-of-the-art autoregressive methods and generates more interpretable and diverse mathematical expressions.
【6】Multi-Model Synthetic Training for Mission-Critical Small Language Models
标题：任务关键小语言模型的多模型综合训练
链接：https://arxiv.org/abs/2509.13047
备注：8 pages. Accepted as a full paper to the 3rd International Conference on Foundation and Large Language Models (IEEE FLLM) 2025
摘要：大型语言模型（LLM）在许多领域都表现出了卓越的能力，但它们在专业领域的应用仍然受到特定领域训练数据的稀缺性和复杂性的限制。我们提出了一种新的方法，通过使用LLM作为一次性教师，而不是直接使用它们进行推理，实现了海上情报成本降低261倍。我们的方法通过多模型生成（GPT-4 o和o3-mini）将32亿条自动识别系统（AIS）船舶跟踪记录转换为21，543个合成问题和答案对，防止过度拟合并确保准确推理。由此产生的微调Qwen2.5- 7 B模型在海上任务中实现了75%的准确率，同时比使用更大的模型进行推理要便宜得多。我们表明，更小，更便宜的模型-当微调适当-可以提供类似的精度相比，更大的模型是昂贵的。我们的工作有助于为专业AI应用程序生成合成数据集的不断增长的领域，并为手动注释不可行的领域提供了一个高度可重复的框架。除了在不断发展的专业小语言模型领域扩大研究之外，我们的方法还可以立即应用于各个行业的海上安全，安全运营和船舶交通管理系统。
摘要：Large Language Models (LLMs) have demonstrated remarkable capabilities across many domains, yet their appli- cation to specialized fields remains constrained by the scarcity and complexity of domain-specific training data. We present a novel approach that achieves a 261x cost reduction for maritime intelligence by using LLMs as one-time teachers rather than using them directly for inference. Our method transforms 3.2 billion Automatic Identification System (AIS) vessel tracking records into 21,543 synthetic question and answer pairs through multi-model generation (GPT-4o and o3-mini), preventing over- fitting and ensuring accurate reasoning. The resulting fine-tuned Qwen2.5-7B model achieves 75% accuracy on maritime tasks, while being substantially cheaper than using a larger model for inference. We show that smaller, cheaper models - when fine tuned properly - can provide similar accuracy compared to larger models that are prohibitively expensive. Our work contributes to the growing field of synthetic dataset generation for specialized AI applications and presents a highly reproducible framework for domains where manual annotation is infeasible. Beyond expand- ing research in the growing field of specialized small language models, our approach has immediate applications in maritime safety, security operations, and vessel traffic management systems in various industries.
【7】Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model
标题：面向对象所有权理解的主动问题生成：大语言模型和概率生成模型
链接：https://arxiv.org/abs/2509.12754
备注：Submitted to AROB-ISBC 2026 (Journal Track option)
摘要：在家庭和办公室环境中工作的机器人必须理解对象所有权，以正确执行诸如“把我的杯子拿来”之类的指令。然而，仅从视觉特征无法可靠地推断所有权。为了解决这一差距，我们提出了主动所有权学习（ActOwL），这是一个框架，使机器人能够主动生成并向用户提出与所有权相关的问题。ActOwL采用概率生成模型来选择最大化信息增益的问题，从而有效地获取所有权知识以提高学习效率。此外，通过利用来自大型语言模型（LLM）的常识知识，对象被预先分类为共享或拥有，并且只有拥有的对象才是提问的目标。通过在模拟家庭环境和真实实验室环境中的实验，ActOwL比基线方法实现了更高的所有权聚类准确性，问题更少。这些发现证明了将主动推理与LLM引导的常识推理相结合的有效性，提高了机器人获得所有权知识以执行实用和社会适当任务的能力。
摘要：Robots operating in domestic and office environments must understand object ownership to correctly execute instructions such as ``Bring me my cup.'' However, ownership cannot be reliably inferred from visual features alone. To address this gap, we propose Active Ownership Learning (ActOwL), a framework that enables robots to actively generate and ask ownership-related questions to users. ActOwL employs a probabilistic generative model to select questions that maximize information gain, thereby acquiring ownership knowledge efficiently to improve learning efficiency. Additionally, by leveraging commonsense knowledge from Large Language Models (LLM), objects are pre-classified as either shared or owned, and only owned objects are targeted for questioning. Through experiments in a simulated home environment and a real-world laboratory setting, ActOwL achieved significantly higher ownership clustering accuracy with fewer questions than baseline methods. These findings demonstrate the effectiveness of combining active inference with LLM-guided commonsense reasoning, advancing the capability of robots to acquire ownership knowledge for practical and socially appropriate task execution.
【8】Large Language Model Scaling Laws for Neural Quantum States in Quantum Chemistry
标题：量子化学中神经量子状态的大语言模型标度定律
链接：https://arxiv.org/abs/2509.12679
备注：16 pages, 5 figures, to be submitted for peer review
摘要：缩放定律已经被用来描述大型语言模型（LLM）的性能如何随着模型大小、训练数据大小或计算资源量而缩放。受神经量子态（NQS）越来越多地采用基于LLM的组件的事实的启发，我们试图了解NQS缩放定律，从而揭示NQS ansatze的可扩展性和最佳性能-资源权衡。特别是，我们确定的比例法则，预测的性能，测量的绝对误差和V-分数，为基于变压器的NQS作为问题的大小的函数，在第二个量子化的量子化学应用。通过对所获得的参数曲线进行类似的计算约束优化，我们发现模型大小和训练时间之间的关系高度依赖于损失度量和方差，并且不遵循语言模型的近似线性关系。
摘要：Scaling laws have been used to describe how large language model (LLM) performance scales with model size, training data size, or amount of computational resources. Motivated by the fact that neural quantum states (NQS) has increasingly adopted LLM-based components, we seek to understand NQS scaling laws, thereby shedding light on the scalability and optimal performance--resource trade-offs of NQS ansatze. In particular, we identify scaling laws that predict the performance, as measured by absolute error and V-score, for transformer-based NQS as a function of problem size in second-quantized quantum chemistry applications. By performing analogous compute-constrained optimization of the obtained parametric curves, we find that the relationship between model size and training time is highly dependent on loss metric and ansatz, and does not follow the approximately linear relationship found for language models.
【9】Instance-level Randomization: Toward More Stable LLM Evaluations
标题：实例级随机化：迈向更稳定的LLM评估
链接：https://arxiv.org/abs/2509.12678
备注：Accepted by Findings of EMNLP 2025
摘要：大型语言模型（LLM）的评估存在不稳定性，其中随机因素（例如Few-Shot示例）的微小变化可能导致分数甚至模型排名的剧烈波动。此外，不同的LLM可以对随机因子的特定设置具有不同的偏好。因此，使用随机因素的固定设置，这通常被采用作为当前评估的范例，可能会导致LLM之间的潜在不公平比较。为了减轻评价的波动性，我们首先从理论上分析了随机因素变化引起的方差的来源。针对这些特定的源，我们提出了实例级随机化（ILR）方法，以减少方差，提高模型比较的公平性。我们没有在单个实验中对整个基准测试使用固定的设置，而是将影响每个实例评估分数的所有因素随机化，运行多个实验并报告平均分数。理论分析和实验结果表明，ILR方法可以有效地降低随机因素引起的方差和不公平比较，并以不到以往方法一半的计算量达到相似的鲁棒性水平。
摘要：Evaluations of large language models (LLMs) suffer from instability, where small changes of random factors such as few-shot examples can lead to drastic fluctuations of scores and even model rankings. Moreover, different LLMs can have different preferences for a certain setting of random factors. As a result, using a fixed setting of random factors, which is often adopted as the paradigm of current evaluations, can lead to potential unfair comparisons between LLMs. To mitigate the volatility of evaluations, we first theoretically analyze the sources of variance induced by changes in random factors. Targeting these specific sources, we then propose the instance-level randomization (ILR) method to reduce variance and enhance fairness in model comparisons. Instead of using a fixed setting across the whole benchmark in a single experiment, we randomize all factors that affect evaluation scores for every single instance, run multiple experiments and report the averaged score. Theoretical analyses and empirical results demonstrate that ILR can reduce the variance and unfair comparisons caused by random factors, as well as achieve similar robustness level with less than half computational cost compared with previous methods.
【10】ScaleDoc: Scaling LLM-based Predicates over Large Document Collections
标题：ScaleDoc：在大型文档集合上扩展基于LLM的预测
链接：https://arxiv.org/abs/2509.12610
摘要：谓词是数据分析系统中的基础组件。然而，现代工作负载越来越多地涉及非结构化文档，这需要语义理解，而不是传统的基于值的谓词。大型语言模型（LLM）在处理大量文档和特定查询时，虽然表现出强大的zero-shot能力，但其高推理成本会导致不可接受的开销。因此，我们引入\textsc{ScaleDoc}，一个新的系统，解决了这个问题，通过解耦谓词执行到离线表示阶段和优化的在线过滤阶段。在离线阶段，\textsc{ScaleDoc}利用LLM为每个文档生成语义表示。在线上，对于每个查询，它都会在这些表示上训练一个轻量级的代理模型来过滤大多数文档，只将模糊的情况转发给LLM进行最终决策。此外，\textsc{ScaleDoc}提出了两项核心创新以实现显著的效率：（1）基于对比学习的框架，用于训练代理模型以生成可靠的预测决策分数;（2）自适应级联机制，用于确定有效的过滤策略，同时满足特定的准确性目标。我们对三个数据集的评估表明，\textsc{ScaleDoc}实现了超过2$\times$的端到端加速，并将昂贵的LLM调用减少了高达85%，使大规模语义分析变得实用和高效。
摘要 ：Predicates are foundational components in data analysis systems. However, modern workloads increasingly involve unstructured documents, which demands semantic understanding, beyond traditional value-based predicates. Given enormous documents and ad-hoc queries, while Large Language Models (LLMs) demonstrate powerful zero-shot capabilities, their high inference cost leads to unacceptable overhead. Therefore, we introduce \textsc{ScaleDoc}, a novel system that addresses this by decoupling predicate execution into an offline representation phase and an optimized online filtering phase. In the offline phase, \textsc{ScaleDoc} leverages a LLM to generate semantic representations for each document. Online, for each query, it trains a lightweight proxy model on these representations to filter the majority of documents, forwarding only the ambiguous cases to the LLM for final decision. Furthermore, \textsc{ScaleDoc} proposes two core innovations to achieve significant efficiency: (1) a contrastive-learning-based framework that trains the proxy model to generate reliable predicating decision scores; (2) an adaptive cascade mechanism that determines the effective filtering policy while meeting specific accuracy targets. Our evaluations across three datasets demonstrate that \textsc{ScaleDoc} achieves over a 2$\times$ end-to-end speedup and reduces expensive LLM invocations by up to 85\%, making large-scale semantic analysis practical and efficient.
【11】Selective Risk Certification for LLM Outputs via Information-Lift Statistics: PAC-Bayes, Robustness, and Skeleton Design
标题：通过信息提升统计对LLM输出进行选择性风险认证：PAC-Bayes、稳健性和骨架设计
链接：https://arxiv.org/abs/2509.12527
摘要：大型语言模型通常会产生看似合理但不正确的输出。像HallBayes这样的现有算法缺乏正式的保证。我们发展了第一个选择分类下的信息提升证书综合理论。我们的贡献是：（一）PAC-贝叶斯\emdash {子伽马}分析扩展超出标准的伯恩斯坦界限;（二）明确的骨架灵敏度定理量化的鲁棒性误指定;（三）故障模式的保证下的假设违反;和（四）一个原则的变分方法骨架建设。在六个数据集和多个模型系列中，我们根据经验验证了假设，在相同的风险下将故障率降低了12- 15\ %，并将运行时开销保持在20\%以下（通过故障率进一步降低）。
摘要：Large language models often produce plausible but incorrect outputs. Existing heuristics such as HallBayes lack formal guarantees. We develop the first comprehensive theory of \emph{information-lift certificates} under selective classification. Our contributions are: (i) a PAC-Bayes \emph{sub-gamma} analysis extending beyond standard Bernstein bounds; (ii) explicit skeleton sensitivity theorems quantifying robustness to misspecification; (iii) failure-mode guarantees under assumption violations; and (iv) a principled variational method for skeleton construction. Across six datasets and multiple model families, we validate assumptions empirically, reduce abstention by 12--15\% at the same risk, and maintain runtime overhead below 20\% (further reduced via batching).
【12】Phi: Preference Hijacking in Multi-modal Large Language Models at Inference Time
标题：Phi：推理时多模式大型语言模型中的偏好劫持
链接：https://arxiv.org/abs/2509.12521
摘要：最近，多模态大型语言模型（MLLM）在各个领域都受到了极大的关注。然而，它们的广泛采用也引起了严重的安全问题。在本文中，我们发现了一个新的安全风险的MLLM：输出偏好的MLLM可以任意操纵精心优化的图像。这种攻击往往会产生与背景相关但有偏见的反应，这些反应既没有明显的伤害，也没有不道德的行为，因此很难被发现。具体来说，我们介绍了一种新的方法，偏好劫持（披），用于操纵MLLM响应偏好使用的偏好劫持图像。我们的方法在推理时工作，不需要修改模型。此外，我们引入了一个通用的劫持扰动-一个可转移的组件，可以嵌入到不同的图像劫持MLLM响应对任何攻击者指定的偏好。不同任务的实验结果证明了我们方法的有效性。Phi的代码可在https://github.com/Yifan-Lan/Phi上访问。
摘要：Recently, Multimodal Large Language Models (MLLMs) have gained significant attention across various domains. However, their widespread adoption has also raised serious safety concerns. In this paper, we uncover a new safety risk of MLLMs: the output preference of MLLMs can be arbitrarily manipulated by carefully optimized images. Such attacks often generate contextually relevant yet biased responses that are neither overtly harmful nor unethical, making them difficult to detect. Specifically, we introduce a novel method, Preference Hijacking (Phi), for manipulating the MLLM response preferences using a preference hijacked image. Our method works at inference time and requires no model modifications. Additionally, we introduce a universal hijacking perturbation -- a transferable component that can be embedded into different images to hijack MLLM responses toward any attacker-specified preferences. Experimental results across various tasks demonstrate the effectiveness of our approach. The code for Phi is accessible at https://github.com/Yifan-Lan/Phi.
【13】SENTRA: Selected-Next-Token Transformer for LLM Text Detection
标题：SENTRA：用于LLM文本检测的选定下一个令牌Transformer
链接：https://arxiv.org/abs/2509.12385
备注：EMNLP Findings 2025
摘要：LLM正变得越来越强大和广泛。因此，其滥用的可能性和现实性也在增加。在这项工作中，我们解决的问题，检测LLM生成的文本，没有明确声明。我们提出了一种新的，通用的，监督LLM文本检测器，选择下一个令牌Transformer（SENTRA）。SENTRA是一个基于transformer的编码器，它利用选择的下一个标记概率序列，并利用大量未标记数据的对比预训练。我们在24个文本域的三个流行的公共数据集上进行的实验表明，SENTRA是一种通用分类器，在域外设置中显著优于流行的基线。
摘要：LLMs are becoming increasingly capable and widespread. Consequently, the potential and reality of their misuse is also growing. In this work, we address the problem of detecting LLM-generated text that is not explicitly declared as such. We present a novel, general-purpose, and supervised LLM text detector, SElected-Next-Token tRAnsformer (SENTRA). SENTRA is a Transformer-based encoder leveraging selected-next-token-probability sequences and utilizing contrastive pre-training on large amounts of unlabeled data. Our experiments on three popular public datasets across 24 domains of text demonstrate SENTRA is a general-purpose classifier that significantly outperforms popular baselines in the out-of-domain setting.
【14】LLMAP: LLM-Assisted Multi-Objective Route Planning with User Preferences
标题：LLMAP：基于用户偏好的LLM辅助多目标路径规划
链接：https://arxiv.org/abs/2509.12273
摘要：大型语言模型（LLM）的兴起使自然语言驱动的路线规划成为一个新兴的研究领域，其中包含丰富的用户目标。目前的研究表现出两种不同的方法：直接路线规划使用LLM作为代理和基于图形的搜索策略。然而，LLM在前一种方法中难以处理大量的地图数据，而后者在理解自然语言偏好方面表现出有限的能力。此外，一个更关键的挑战来自全球用户的高度异质性和不可预测的时空分布。在本文中，我们介绍了一种新的LLM-Assisted路线规划（LLMAP）系统，该系统采用LLM-as-Parser来理解自然语言，识别任务，提取用户偏好并识别任务依赖关系，再加上多步图构造迭代搜索（MSGS）算法作为最佳路线查找的底层求解器。我们的多目标优化方法自适应地调整目标权重，以最大限度地提高兴趣点（POI）的质量和任务完成率，同时最大限度地减少路线距离，受三个关键约束：用户的时间限制，POI开放时间和任务的依赖性。我们使用1,000个路由提示进行了广泛的实验，这些提示在全球14个国家和27个城市中以不同的复杂性进行采样。结果表明，我们的方法实现了优越的性能，保证在多个约束。
摘要 ：The rise of large language models (LLMs) has made natural language-driven route planning an emerging research area that encompasses rich user objectives. Current research exhibits two distinct approaches: direct route planning using LLM-as-Agent and graph-based searching strategies. However, LLMs in the former approach struggle to handle extensive map data, while the latter shows limited capability in understanding natural language preferences. Additionally, a more critical challenge arises from the highly heterogeneous and unpredictable spatio-temporal distribution of users across the globe. In this paper, we introduce a novel LLM-Assisted route Planning (LLMAP) system that employs an LLM-as-Parser to comprehend natural language, identify tasks, and extract user preferences and recognize task dependencies, coupled with a Multi-Step Graph construction with iterative Search (MSGS) algorithm as the underlying solver for optimal route finding. Our multi-objective optimization approach adaptively tunes objective weights to maximize points of interest (POI) quality and task completion rate while minimizing route distance, subject to three key constraints: user time limits, POI opening hours, and task dependencies. We conduct extensive experiments using 1,000 routing prompts sampled with varying complexity across 14 countries and 27 cities worldwide. The results demonstrate that our approach achieves superior performance with guarantees across multiple constraints.
【15】MEUV: Achieving Fine-Grained Capability Activation in Large Language Models via Mutually Exclusive Unlock Vectors
标题：MEV：通过相互排斥的数据库Vectors在大型语言模型中实现细粒度的能力激活
链接：https://arxiv.org/abs/2509.12221
备注：Under Review
摘要：大型语言模型（LLM）强制执行安全对齐以可靠地拒绝恶意请求，但同样的全面保护措施也会阻止在警务，国防和其他高风险环境中的合法使用。早期的“参照方向”编辑可以绕过这些层，但它们依赖于一个单一的向量，不加区别地解锁所有危险的主题，不提供语义控制。我们引入互斥拒绝向量（MEUV），一个轻量级的框架，分解成主题对齐的，几乎正交的向量，每个专用于一个敏感的能力的单片拒绝方向。MEUV是在一个单一的时代学习的多任务目标，融合了差分消融裕度，交叉主题和正交性的处罚，以及几个辅助条款。在双语恶意提示基准测试中，MEUV在Gemma-2-2B、LLaMA-3-8B和Qwen-7 B上的攻击成功率不低于87%，但与最佳单向基准相比，跨主题泄漏减少了90%。在中文训练的向量几乎不变地转移到英语（反之亦然），这表明语言不可知的拒绝子空间。结果表明，细粒度的，主题级的能力激活是可以实现的，最小的效用损失，铺平了道路，在安全敏感领域的受控LLM部署。
摘要：Large language models (LLMs) enforce safety alignment to reliably refuse malicious requests, yet the same blanket safeguards also block legitimate uses in policing, defense, and other high-stakes settings. Earlier "refusal-direction" edits can bypass those layers, but they rely on a single vector that indiscriminately unlocks all hazardous topics, offering no semantic control. We introduce Mutually Exclusive Unlock Vectors (MEUV), a lightweight framework that factorizes the monolithic refusal direction into topic-aligned, nearly orthogonal vectors, each dedicated to one sensitive capability. MEUV is learned in a single epoch with a multi-task objective that blends a differential-ablation margin, cross-topic and orthogonality penalties, and several auxiliary terms. On bilingual malicious-prompt benchmarks, MEUV achieves an attack success rate of no less than 87% on Gemma-2-2B, LLaMA-3-8B, and Qwen-7B, yet cuts cross-topic leakage by up to 90% compared with the best single-direction baseline. Vectors trained in Chinese transfer almost unchanged to English (and vice versa), suggesting a language-agnostic refusal subspace. The results show that fine-grained, topic-level capability activation is achievable with minimal utility loss, paving the way for controlled LLMs deployment in security-sensitive domains.
Graph相关(图学习|图神经网络|图优化等)(10篇)
【1】B-TGAT: A Bi-directional Temporal Graph Attention Transformer for Clustering Multivariate Spatiotemporal Data
标题：B-TGAT：用于对多元时空数据进行聚集的双向时态图注意力Transformer
链接：https://arxiv.org/abs/2509.13202
备注：10 pages, In review
摘要：高维多变量时空气候数据的聚类是具有挑战性的，由于复杂的时间依赖性，不断变化的空间相互作用，和非平稳动力学。传统的聚类方法，包括循环和卷积模型，通常难以在保留空间上下文的同时捕获局部和全局时间关系。我们提出了一个时间分布式的混合U-Net自动编码器，它集成了一个双向时间图注意力Transformer（B-TGAT），以指导多维时空气候数据集的有效时间聚类。编码器和解码器配备了ConvLSTM 2D模块，通过对局部动态和随时间变化的空间相关性进行建模来提取联合时空特征，并跳过在特征压缩和重建过程中保留多尺度空间细节的连接。在瓶颈处，B-TGAT将基于图的空间建模与注意力驱动的时间编码相结合，实现了时间邻居的自适应加权，并捕获了跨区域的短期和长期依赖关系。该架构产生针对聚类优化的有区别的潜在嵌入。在三个不同的时空气候数据集上的实验表明，与最先进的基线相比，具有优越的聚类分离性，时间稳定性和与已知气候转变的一致性。ConvLSTM 2D、U-Net skip connections和B-TGAT的集成增强了时间聚类性能，同时为复杂的时空变化提供了可解释的见解，推动了方法的发展和气候科学的应用。
摘要：Clustering high-dimensional multivariate spatiotemporal climate data is challenging due to complex temporal dependencies, evolving spatial interactions, and non-stationary dynamics. Conventional clustering methods, including recurrent and convolutional models, often struggle to capture both local and global temporal relationships while preserving spatial context. We present a time-distributed hybrid U-Net autoencoder that integrates a Bi-directional Temporal Graph Attention Transformer (B-TGAT) to guide efficient temporal clustering of multidimensional spatiotemporal climate datasets. The encoder and decoder are equipped with ConvLSTM2D modules that extract joint spatial--temporal features by modeling localized dynamics and spatial correlations over time, and skip connections that preserve multiscale spatial details during feature compression and reconstruction. At the bottleneck, B-TGAT integrates graph-based spatial modeling with attention-driven temporal encoding, enabling adaptive weighting of temporal neighbors and capturing both short and long-range dependencies across regions. This architecture produces discriminative latent embeddings optimized for clustering. Experiments on three distinct spatiotemporal climate datasets demonstrate superior cluster separability, temporal stability, and alignment with known climate transitions compared to state-of-the-art baselines. The integration of ConvLSTM2D, U-Net skip connections, and B-TGAT enhances temporal clustering performance while providing interpretable insights into complex spatiotemporal variability, advancing both methodological development and climate science applications.
【2】Learning from Heterophilic Graphs: A Spectral Theory Perspective on the Impact of Self-Loops and Parallel Edges
标题：从异嗜图中学习：谱理论视角研究自循环和平行边的影响
链接：https://arxiv.org/abs/2509.13139
摘要：图的异质性对消息传递图神经网络（MP-GNNs）的性能提出了严峻的挑战。像图卷积网络（GCN）这样熟悉的低通滤波器面临性能下降，这可以归因于来自不同相邻节点的消息的混合。低通滤波器在异嗜图上的性能仍然需要深入的分析。在这种情况下，我们通过添加一些自环和平行边来更新heterophilic图。我们观察到，图拉普拉斯算子的特征值分别减少和增加的自环和平行边的数量。我们进行了几项研究GCN的性能在各种基准heterophilic网络通过添加自循环或平行的边缘。研究表明，GCN表现出增加或减少的性能趋势增加自循环和平行的边缘。在此基础上，我们建立了异嗜图上的图谱与低通滤波器性能趋势之间的联系。图谱表征输入图的基本内在属性，如连通组件，稀疏性，平均度，聚类结构等的存在。我们的工作是善于无缝评估图谱和性能，通过观察低通滤波器的性能趋势，而不追求昂贵的特征值分解。理论基础也进行了讨论，以验证添加自环和平行边的图的谱的影响。
摘要 ：Graph heterophily poses a formidable challenge to the performance of Message-passing Graph Neural Networks (MP-GNNs). The familiar low-pass filters like Graph Convolutional Networks (GCNs) face performance degradation, which can be attributed to the blending of the messages from dissimilar neighboring nodes. The performance of the low-pass filters on heterophilic graphs still requires an in-depth analysis. In this context, we update the heterophilic graphs by adding a number of self-loops and parallel edges. We observe that eigenvalues of the graph Laplacian decrease and increase respectively by increasing the number of self-loops and parallel edges. We conduct several studies regarding the performance of GCN on various benchmark heterophilic networks by adding either self-loops or parallel edges. The studies reveal that the GCN exhibited either increasing or decreasing performance trends on adding self-loops and parallel edges. In light of the studies, we established connections between the graph spectra and the performance trends of the low-pass filters on the heterophilic graphs. The graph spectra characterize the essential intrinsic properties of the input graph like the presence of connected components, sparsity, average degree, cluster structures, etc. Our work is adept at seamlessly evaluating graph spectrum and properties by observing the performance trends of the low-pass filters without pursuing the costly eigenvalue decomposition. The theoretical foundations are also discussed to validate the impact of adding self-loops and parallel edges on the graph spectrum.
【3】Spatiotemporal graph neural process for reconstruction, extrapolation, and classification of cardiac trajectories
标题：用于心脏轨迹重建、外推和分类的时空图神经过程
链接：https://arxiv.org/abs/2509.12953
摘要：我们提出了一个概率框架，从稀疏的观测结构化时空动态建模，专注于心脏运动。我们的方法将神经常微分方程（NODE），图形神经网络（GNNs）和神经过程集成到一个统一的模型中，该模型可以捕获不确定性，时间连续性和解剖结构。我们将动态系统表示为时空复用图，并使用GNN参数化向量场对其潜在轨迹进行建模。给定节点和边缘级别的稀疏上下文观测，该模型推断潜在初始状态和控制变量的分布，从而实现轨迹的内插和外推。我们在三个合成动力学系统（耦合摆，洛伦兹吸引子和仓本振荡器）和两个真实世界的心脏成像数据集ACDC（N=150）和英国生物银行（N=526）上验证了该方法-展示了准确的重建，外推和疾病分类能力。该模型准确地重建轨迹，并从单个观察到的周期推断未来的心动周期。它在ACDC分类任务上实现了最先进的结果（高达99%的准确率），并以具有竞争力的性能（高达67%的准确率）检测英国生物库受试者的房颤。这项工作介绍了一种灵活的方法来分析心脏运动，并提供了基于图形的学习结构化生物医学时空时间序列数据的基础。
摘要：We present a probabilistic framework for modeling structured spatiotemporal dynamics from sparse observations, focusing on cardiac motion. Our approach integrates neural ordinary differential equations (NODEs), graph neural networks (GNNs), and neural processes into a unified model that captures uncertainty, temporal continuity, and anatomical structure. We represent dynamic systems as spatiotemporal multiplex graphs and model their latent trajectories using a GNN-parameterized vector field. Given the sparse context observations at node and edge levels, the model infers a distribution over latent initial states and control variables, enabling both interpolation and extrapolation of trajectories. We validate the method on three synthetic dynamical systems (coupled pendulum, Lorenz attractor, and Kuramoto oscillators) and two real-world cardiac imaging datasets - ACDC (N=150) and UK Biobank (N=526) - demonstrating accurate reconstruction, extrapolation, and disease classification capabilities. The model accurately reconstructs trajectories and extrapolates future cardiac cycles from a single observed cycle. It achieves state-of-the-art results on the ACDC classification task (up to 99% accuracy), and detects atrial fibrillation in UK Biobank subjects with competitive performance (up to 67% accuracy). This work introduces a flexible approach for analyzing cardiac motion and offers a foundation for graph-based learning in structured biomedical spatiotemporal time-series data.
【4】A Graph Machine Learning Approach for Detecting Topological Patterns in Transactional Graphs
标题：一种检测跨图中拓扑模式的图机器学习方法
链接：https://arxiv.org/abs/2509.12730
备注：Paper accepted @ Workshop on AI for Financial Crime Fight (AI4FCF @ ICDM 2025)
摘要：数字生态系统的兴起使金融部门面临不断演变的滥用和犯罪策略，这些策略在不同环境（基于法定货币的资产、加密资产等）内和跨环境共享运营知识和技术。传统的基于规则的系统缺乏检测复杂或协调的犯罪行为（模式）所需的适应性，突出了对分析行为人互动以发现可疑活动并提取其作案手法的策略的需求。出于这个原因，在这项工作中，我们提出了一种集成图机器学习和网络分析的方法，以提高对事务图中已知拓扑模式的检测。然而，一个关键的挑战在于传统金融数据集的局限性，这些数据集通常提供稀疏的、未标记的信息，难以用于基于图的模式分析。因此，我们首先提出了一个四步预处理框架，包括（i）提取图结构，（ii）考虑数据的时间性来管理大型节点集，（iii）检测社区，以及（iv）应用自动标记策略来生成弱地面实况标签。然后，一旦处理完数据，就会实现Graph Autoencoders来区分已知的拓扑模式。具体而言，三种不同的GAE变体的实施和比较在此分析。初步结果表明，这种模式为重点，拓扑驱动的方法是有效的检测复杂的金融犯罪计划，提供了一个有前途的替代传统的基于规则的检测系统。
摘要：The rise of digital ecosystems has exposed the financial sector to evolving abuse and criminal tactics that share operational knowledge and techniques both within and across different environments (fiat-based, crypto-assets, etc.). Traditional rule-based systems lack the adaptability needed to detect sophisticated or coordinated criminal behaviors (patterns), highlighting the need for strategies that analyze actors' interactions to uncover suspicious activities and extract their modus operandi. For this reason, in this work, we propose an approach that integrates graph machine learning and network analysis to improve the detection of well-known topological patterns within transactional graphs. However, a key challenge lies in the limitations of traditional financial datasets, which often provide sparse, unlabeled information that is difficult to use for graph-based pattern analysis. Therefore, we firstly propose a four-step preprocessing framework that involves (i) extracting graph structures, (ii) considering data temporality to manage large node sets, (iii) detecting communities within, and (iv) applying automatic labeling strategies to generate weak ground-truth labels. Then, once the data is processed, Graph Autoencoders are implemented to distinguish among the well-known topological patterns. Specifically, three different GAE variants are implemented and compared in this analysis. Preliminary results show that this pattern-focused, topology-driven method is effective for detecting complex financial crime schemes, offering a promising alternative to conventional rule-based detection systems.
【5】Unbiased Online Curvature Approximation for Regularized Graph Continual Learning
标题：正规图连续学习的无偏在线曲线逼近
链接：https://arxiv.org/abs/2509.12727
备注：9 pages
摘要：图持续学习（GCL）旨在从基于图的任务的连续序列中学习。正则化方法对于防止GCL中的灾难性遗忘至关重要，特别是在具有挑战性的无重放，类增量设置中，每个任务都由一组唯一的类组成。在这项工作中，我们首先建立了一个通用的正则化框架的基础上弯曲的参数空间的Fisher信息矩阵（FisherInformation Matrix，简称FMM）。我们表明，占主导地位的弹性权重合并（EWC）及其变体是一个特殊的情况下，在这个框架内，使用对角近似的经验的基于参数从以前的任务。为了克服它们的局限性，我们提出了一个新的无偏在线曲率近似的完整的基于模型的当前学习状态。我们的方法直接估计的正则化项在一个在线的方式，而不显式地评估和存储的正则化本身。这使得模型能够在学习新任务的过程中更好地捕捉损失情况，同时保留从以前的任务中学到的知识。在三个图数据集上的大量实验表明，我们的方法显著优于现有的基于正则化的方法，在稳定性（保留旧知识）和可塑性（获取新知识）之间实现了更好的权衡。
摘要：Graph continual learning (GCL) aims to learn from a continuous sequence of graph-based tasks. Regularization methods are vital for preventing catastrophic forgetting in GCL, particularly in the challenging replay-free, class-incremental setting, where each task consists of a set of unique classes. In this work, we first establish a general regularization framework for GCL based on the curved parameter space induced by the Fisher information matrix (FIM). We show that the dominant Elastic Weight Consolidation (EWC) and its variants are a special case within this framework, using a diagonal approximation of the empirical FIM based on parameters from previous tasks. To overcome their limitations, we propose a new unbiased online curvature approximation of the full FIM based on the model's current learning state. Our method directly estimates the regularization term in an online manner without explicitly evaluating and storing the FIM itself. This enables the model to better capture the loss landscape during learning new tasks while retaining the knowledge learned from previous tasks. Extensive experiments on three graph datasets demonstrate that our method significantly outperforms existing regularization-based methods, achieving a superior trade-off between stability (retaining old knowledge) and plasticity (acquiring new knowledge).
【6】Soft Graph Transformer for MIMO Detection
标题：用于多输入多输出检测的软图Transformer
链接
：https://arxiv.org/abs/2509.12694
备注：8 pages
摘要：我们提出了软图Transformer（SGT），一个软输入软输出的MIMO检测的神经架构。虽然最大似然（ML）检测达到了最佳的精度，但其令人望而却步的指数复杂性使其对现实世界的系统不切实际。传统的消息传递算法提供了易于处理的替代方案，但依赖于大系统渐近性和随机矩阵假设，这两者在实际实现中都失败了。另一方面，现有的基于变换器的检测器未能结合MIMO因子图结构，并且不能利用解码器侧软信息，从而限制了它们的独立性能和它们在迭代检测解码（IDD）中的适用性。为了克服这些限制，SGT将消息传递直接集成到图形感知注意力机制中，并通过软输入嵌入支持解码器通知更新。这种设计能够在保持计算效率的同时实现有效的软输出生成。作为一个独立的检测器，SGT非常接近ML性能，并超过了之前基于transformer的方法。
摘要：We propose the Soft Graph Transformer (SGT), a Soft-Input-Soft-Output neural architecture tailored for MIMO detection. While Maximum Likelihood (ML) detection achieves optimal accuracy, its prohibitive exponential complexity renders it impractical for real-world systems. Conventional message passing algorithms offer tractable alternatives but rely on large-system asymptotics and random matrix assumptions, both of which break down under practical implementations. Prior Transformer-based detectors, on the other hand, fail to incorporate the MIMO factor graph structure and cannot utilize decoder-side soft information, limiting their standalone performance and their applicability in iterative detection-decoding (IDD). To overcome these limitations, SGT integrates message passing directly into a graph-aware attention mechanism and supports decoder-informed updates through soft-input embeddings. This design enables effective soft-output generation while preserving computational efficiency. As a standalone detector, SGT closely approaches ML performance and surpasses prior Transformer-based approaches.
【7】Graph Homophily Booster: Rethinking the Role of Discrete Features on Heterophilic Graphs
标题：图同质性助推器：重新思考离散特征在异同质图上的作用
链接：https://arxiv.org/abs/2509.12530
备注：14 pages
摘要：图神经网络（GNN）已经成为对图结构数据建模的强大工具。然而，现有的GNN经常与异嗜图（heterophilic graphs）作斗争，其中连接的节点往往具有不同的特征或标签。虽然已经提出了许多方法来解决这一挑战，但它们主要集中在架构设计上，而没有直接针对异质性问题的根本原因。这些方法在具有挑战性的异嗜性数据集上的表现甚至比最简单的MLP更差。例如，我们的实验表明，21个最新的GNN仍然落后于Actor数据集上的MLP。这一关键挑战需要一种创新的方法来解决架构设计之外的图形异质性。为了弥合这一差距，我们提出并研究了一个新的和未探索的范式：直接增加通过精心设计的图变换的图同伦。在这项工作中，我们提出了一个简单而有效的框架，称为GRAPHITE，以解决图形异质性。据我们所知，这项工作是第一个方法，显式地转换图直接改善图的同质性。源于同质性的确切定义，我们提出的GRAPHITE创建特征节点，以促进共享相似特征的节点之间的同质性消息传递。此外，我们从理论上和经验表明，我们提出的GRAPHITE显着增加了原本heterophilic图的同质性，只有轻微增加的图形大小。在具有挑战性的数据集上进行的大量实验表明，我们提出的GRAPHITE在异嗜图上的性能明显优于最先进的方法，同时在同嗜图上实现了与最先进的方法相当的准确性。
摘要：Graph neural networks (GNNs) have emerged as a powerful tool for modeling graph-structured data. However, existing GNNs often struggle with heterophilic graphs, where connected nodes tend to have dissimilar features or labels. While numerous methods have been proposed to address this challenge, they primarily focus on architectural designs without directly targeting the root cause of the heterophily problem. These approaches still perform even worse than the simplest MLPs on challenging heterophilic datasets. For instance, our experiments show that 21 latest GNNs still fall behind the MLP on the Actor dataset. This critical challenge calls for an innovative approach to addressing graph heterophily beyond architectural designs. To bridge this gap, we propose and study a new and unexplored paradigm: directly increasing the graph homophily via a carefully designed graph transformation. In this work, we present a simple yet effective framework called GRAPHITE to address graph heterophily. To the best of our knowledge, this work is the first method that explicitly transforms the graph to directly improve the graph homophily. Stemmed from the exact definition of homophily, our proposed GRAPHITE creates feature nodes to facilitate homophilic message passing between nodes that share similar features. Furthermore, we both theoretically and empirically show that our proposed GRAPHITE significantly increases the homophily of originally heterophilic graphs, with only a slight increase in the graph size. Extensive experiments on challenging datasets demonstrate that our proposed GRAPHITE significantly outperforms state-of-the-art methods on heterophilic graphs while achieving comparable accuracy with state-of-the-art methods on homophilic graphs.
【8】Finite-Agent Stochastic Differential Games on Large Graphs: II. Graph-Based Architectures
标题：大图上的代理-代理随机微博弈：II。基于图形的架构
链接：https://arxiv.org/abs/2509.12484
摘要：我们提出了一种新的神经网络架构，称为不可训练的修改（NTM），用于计算图上的随机微分游戏（SDG）中的纳什均衡。这些游戏模拟了金融、机器人、能源和社会动态中出现的广泛的图形结构多智能体系统，其中智能体在不确定性下进行局部交互。NTM架构在前馈神经网络上施加了一种图引导的稀疏化，嵌入了与底层图拓扑对齐的固定的、不可训练的组件。这种设计增强了可解释性和稳定性，同时显著减少了大规模稀疏设置中可训练参数的数量。我们从理论上建立了一个通用的近似性质的NTM在静态游戏的图和数值验证其表现力和鲁棒性，通过监督学习任务。在此基础上，我们将NTM整合到两个最先进的游戏求解器中，直接参数化和深度Boundary，产生它们的稀疏变体（NTM-DP和NTM-DBoundary）。在不同图结构的三个SDG上进行的数值实验表明，基于NTM的方法实现了与完全可训练的方法相当的性能，同时提高了计算效率。
摘要：We propose a novel neural network architecture, called Non-Trainable Modification (NTM), for computing Nash equilibria in stochastic differential games (SDGs) on graphs. These games model a broad class of graph-structured multi-agent systems arising in finance, robotics, energy, and social dynamics, where agents interact locally under uncertainty. The NTM architecture imposes a graph-guided sparsification on feedforward neural networks, embedding fixed, non-trainable components aligned with the underlying graph topology. This design enhances interpretability and stability, while significantly reducing the number of trainable parameters in large-scale, sparse settings. We theoretically establish a universal approximation property for NTM in static games on graphs and numerically validate its expressivity and robustness through supervised learning tasks. Building on this foundation, we incorporate NTM into two state-of-the-art game solvers, Direct Parameterization and Deep BSDE, yielding their sparse variants (NTM-DP and NTM-DBSDE). Numerical experiments on three SDGs across various graph structures demonstrate that NTM-based methods achieve performance comparable to their fully trainable counterparts, while offering improved computational efficiency.
【9】Unsupervised Atomic Data Mining via Multi-Kernel Graph Autoencoders for Machine Learning Force Fields
标题：基于多核图自编码器的机器学习力场无监督原子数据挖掘
链接：https://arxiv.org/abs/2509.12358
摘要 ：在避免采样偏差的同时构建化学多样性数据集对于训练有效和可推广的力场至关重要。然而，在计算化学和材料科学中，许多常见的数据集生成技术容易对势能面的区域进行过采样。此外，这些区域可能难以识别和相互隔离，或者可能与人类直觉不一致，这使得系统地消除数据集中的偏差具有挑战性。虽然传统的聚类和修剪（下采样）方法可能对此有用，但由于与原子描述符的高维性相关的困难，它们通常会导致信息丢失或无法正确识别势能表面的不同区域。在这项工作中，我们介绍了基于多核边缘注意力的图形自动编码器（MEAGESTO）模型，这是一种用于分析原子数据集的无监督方法。MEAGrandom将多个线性内核转换与基于注意力的消息传递相结合，以捕获几何敏感性，并在不依赖标签或大量训练的情况下实现有效的数据集修剪。在铌、钽和铁数据集上的示范应用表明，MEAGlets有效地将类似的原子环境分组，允许使用基本的修剪技术来消除采样偏差。该方法为表示学习和聚类提供了一种有效的方法，可用于数据分析，离群点检测和数据集优化。
摘要：Constructing a chemically diverse dataset while avoiding sampling bias is critical to training efficient and generalizable force fields. However, in computational chemistry and materials science, many common dataset generation techniques are prone to oversampling regions of the potential energy surface. Furthermore, these regions can be difficult to identify and isolate from each other or may not align well with human intuition, making it challenging to systematically remove bias in the dataset. While traditional clustering and pruning (down-sampling) approaches can be useful for this, they can often lead to information loss or a failure to properly identify distinct regions of the potential energy surface due to difficulties associated with the high dimensionality of atomic descriptors. In this work, we introduce the Multi-kernel Edge Attention-based Graph Autoencoder (MEAGraph) model, an unsupervised approach for analyzing atomic datasets. MEAGraph combines multiple linear kernel transformations with attention-based message passing to capture geometric sensitivity and enable effective dataset pruning without relying on labels or extensive training. Demonstrated applications on niobium, tantalum, and iron datasets show that MEAGraph efficiently groups similar atomic environments, allowing for the use of basic pruning techniques for removing sampling bias. This approach provides an effective method for representation learning and clustering that can be used for data analysis, outlier detection, and dataset optimization.
【10】Quantum-Inspired Stacked Integrated Concept Graph Model (QISICGM) for Diabetes Risk Prediction
标题：用于糖尿病风险预测的量子启发堆叠集成概念图模型（QISICGM）
链接：https://arxiv.org/abs/2509.12259
备注：13 pages, 3 figures, includes performance tables and visualizations. Proposes a Quantum-Inspired Stacked Integrated Concept Graph Model (QISICGM) that integrates phase feature mapping, self-improving concept graphs, and neighborhood sequence modeling within a stacked ensemble. Demonstrates improved F1 and AUC on an augmented PIMA Diabetes dataset with efficient CPU inference
摘要：QISICGM是一个创新的机器学习框架，利用量子启发的技术来预测糖尿病风险，具有卓越的准确性和效率。QISICGM利用PIMA Indians Diabetes数据集，增加了2，000个合成样本，以减轻类别不平衡（总计：2，768个样本，1，949个阳性），将自我改进的概念图与堆叠集成集成，包括随机森林（RF），额外树（ET），Transformers，卷积神经网络（CNN）和前馈神经网络（FFNN）。该方法实现了0.8933的折外（OOF）F1评分和0.8699的AUC，优于传统方法。受量子启发的元素，如相位特征映射和邻域序列建模，丰富了特征表示，实现了每秒8.5行的CPU高效推理。本文介绍了详细的架构、理论基础、代码见解和性能评估，包括outputs子文件夹的可视化。开源实现（v1.0.0）可在https：//github.com/keninayoung/QISICGM上获得，将QISICGM定位为糖尿病及其他领域AI辅助临床分诊的潜在基准。最终，这项工作通过校准、可解释性和开源可重复性强调了值得信赖的人工智能。
摘要：The Quantum-Inspired Stacked Integrated Concept Graph Model (QISICGM) is an innovative machine learning framework that harnesses quantum-inspired techniques to predict diabetes risk with exceptional accuracy and efficiency. Utilizing the PIMA Indians Diabetes dataset augmented with 2,000 synthetic samples to mitigate class imbalance (total: 2,768 samples, 1,949 positives), QISICGM integrates a self-improving concept graph with a stacked ensemble comprising Random Forests (RF), Extra Trees (ET), transformers, convolutional neural networks (CNNs), and feed-forward neural networks (FFNNs). This approach achieves an out-of-fold (OOF) F1 score of 0.8933 and an AUC of 0.8699, outperforming traditional methods. Quantum inspired elements, such as phase feature mapping and neighborhood sequence modeling, enrich feature representations, enabling CPU-efficient inference at 8.5 rows per second. This paper presents a detailed architecture, theoretical foundations, code insights, and performance evaluations, including visualizations from the outputs subfolder. The open-source implementation (v1.0.0) is available at https://github.com/keninayoung/QISICGM, positioning QISICGM as a potential benchmark for AI-assisted clinical triage in diabetes and beyond. Ultimately, this work emphasizes trustworthy AI through calibration, interpretability, and open-source reproducibility.
Transformer(2篇)
【1】BATR-FST: Bi-Level Adaptive Token Refinement for Few-Shot Transformers
标题：BATR-FST：针对Few-ShotTransformer的双级自适应令牌优化
链接：https://arxiv.org/abs/2509.12768
备注：This paper has been accepted for publication at the IEEE International Joint Conference on Neural Networks (IJCNN), Rome, Italy 2025
摘要：Vision Transformers（ViTs）在计算机视觉应用中显示出了巨大的前景。然而，它们在Few-Shot学习中的表现受到了细化令牌级交互、处理有限训练数据以及形成强烈归纳偏差等挑战的限制。现有的方法往往依赖于不灵活的令牌匹配或基本的相似性措施，这限制了有效地结合全局上下文和本地化的功能细化。为了解决这些挑战，我们提出了双层自适应令牌细化的Few-Shot Transformers（BATR-FST），一个两阶段的方法，逐步提高令牌表示，并保持一个强大的归纳偏见的Few-Shot分类。在预训练阶段，掩蔽图像建模（MIM）通过重新创建掩蔽图像区域来为Vision Transformers（ViT）提供可转移的块级表示，从而为后续适应提供稳健的基础。在元微调阶段，BATR-FST集成了一个双层自适应令牌细化模块，该模块利用令牌聚类来捕获本地化的交互，不确定性感知令牌加权来优先考虑可靠的功能，以及一个双层注意力机制来平衡集群内和集群间的关系，从而促进彻底的令牌细化。此外，图令牌传播确保支持和查询实例之间的语义一致性，而类分离惩罚保留不同的类边界，增强了区分能力。在三个基准Few-Shot数据集上的大量实验表明，BATR-FST在1次和5次拍摄场景下都取得了优异的结果，并通过Transformers改进了Few-Shot分类。
摘要：Vision Transformers (ViTs) have shown significant promise in computer vision applications. However, their performance in few-shot learning is limited by challenges in refining token-level interactions, struggling with limited training data, and developing a strong inductive bias. Existing methods often depend on inflexible token matching or basic similarity measures, which limit the effective incorporation of global context and localized feature refinement. To address these challenges, we propose Bi-Level Adaptive Token Refinement for Few-Shot Transformers (BATR-FST), a two-stage approach that progressively improves token representations and maintains a robust inductive bias for few-shot classification. During the pre-training phase, Masked Image Modeling (MIM) provides Vision Transformers (ViTs) with transferable patch-level representations by recreating masked image regions, providing a robust basis for subsequent adaptation. In the meta-fine-tuning phase, BATR-FST incorporates a Bi-Level Adaptive Token Refinement module that utilizes Token Clustering to capture localized interactions, Uncertainty-Aware Token Weighting to prioritize dependable features, and a Bi-Level Attention mechanism to balance intra-cluster and inter-cluster relationships, thereby facilitating thorough token refinement. Furthermore, Graph Token Propagation ensures semantic consistency between support and query instances, while a Class Separation Penalty preserves different class borders, enhancing discriminative capability. Extensive experiments on three benchmark few-shot datasets demonstrate that BATR-FST achieves superior results in both 1-shot and 5-shot scenarios and improves the few-shot classification via transformers.
【2】A comparison of pipelines for the translation of a low resource language based on transformers
标题：基于转换器的低资源语言翻译管道的比较
链接：https://arxiv.org/abs/2509.12514
备注：9 pages, 4 figures
摘要 ：这项工作比较了用于训练基于transformer的神经网络的三个管道，以生产Bambara的机器翻译器，Bambara是非洲约14，188，850人使用的Mand\`e语言。第一条管道训练一个简单的Transformer将法语句子翻译成Bambara。第二个微调LLaMA 3（3B-8B）教师模型使用解码器只架构的法语到班巴拉翻译。来自前两个管道的模型使用不同的超参数组合进行训练，以提高BLEU和chrF分数，并在测试句子和官方Bambara基准上进行评估。第三条管道使用语言蒸馏和学生-教师双神经网络将Bambara集成到预训练的LaBSE模型中，该模型提供与语言无关的嵌入。然后将BERT扩展应用于LaBSE以生成翻译。所有管道都在Dokotoro（医疗）和Bayelemagaba（混合域）上进行了测试。结果表明，第一个管道虽然更简单，但达到了最佳的翻译准确率（Bayelemagaba上的10%BLEU，21%chrF），与低资源翻译结果一致。在为这项工作创建的Yiri数据集上，它实现了33.81%的BLEU和41%的chrF。基于指令的模型在单个数据集上的表现比在聚合集合上的表现更好，这表明它们更有效地捕捉特定于
摘要：This work compares three pipelines for training transformer-based neural networks to produce machine translators for Bambara, a Mand\`e language spoken in Africa by about 14,188,850 people. The first pipeline trains a simple transformer to translate sentences from French into Bambara. The second fine-tunes LLaMA3 (3B-8B) instructor models using decoder-only architectures for French-to-Bambara translation. Models from the first two pipelines were trained with different hyperparameter combinations to improve BLEU and chrF scores, evaluated on both test sentences and official Bambara benchmarks. The third pipeline uses language distillation with a student-teacher dual neural network to integrate Bambara into a pre-trained LaBSE model, which provides language-agnostic embeddings. A BERT extension is then applied to LaBSE to generate translations. All pipelines were tested on Dokotoro (medical) and Bayelemagaba (mixed domains). Results show that the first pipeline, although simpler, achieves the best translation accuracy (10% BLEU, 21% chrF on Bayelemagaba), consistent with low-resource translation results. On the Yiri dataset, created for this work, it achieves 33.81% BLEU and 41% chrF. Instructor-based models perform better on single datasets than on aggregated collections, suggesting they capture dataset-specific patterns more effectively.
GAN|对抗|攻击|生成相关(7篇)
【1】JANUS: A Dual-Constraint Generative Framework for Stealthy Node Injection Attacks
标题：JANUS：一个用于隐形节点注入攻击的双约束生成框架
链接：https://arxiv.org/abs/2509.13266
摘要：图神经网络（GNN）在各种应用中表现出了卓越的性能，但它们容易受到复杂的对抗性攻击，特别是节点注入攻击。此类攻击的成功在很大程度上依赖于它们的隐蔽性，即与原始图形融合并逃避检测的能力。然而，现有的方法往往通过依赖于间接代理度量来实现隐形，缺乏对注入内容的基本特征的考虑，或者只关注于模仿局部结构，这导致了局部近视问题。为了克服这些限制，我们提出了一个双重约束的隐形节点注入框架，称为节点和通用结构的联合对齐（JANUS）。在局部层次上，我们引入了局部特征流形对齐策略，以实现特征空间的几何一致性。在全局层次上，我们将结构化的潜变量和最大化的互信息与生成的结构，确保注入的结构是一致的原始图的语义模式。我们将注入攻击建模为一个顺序决策过程，并通过强化学习代理进行优化。在多个标准数据集上的实验表明，JANUS框架在攻击有效性和隐蔽性方面明显优于现有方法。
摘要：Graph Neural Networks (GNNs) have demonstrated remarkable performance across various applications, yet they are vulnerable to sophisticated adversarial attacks, particularly node injection attacks. The success of such attacks heavily relies on their stealthiness, the ability to blend in with the original graph and evade detection. However, existing methods often achieve stealthiness by relying on indirect proxy metrics, lacking consideration for the fundamental characteristics of the injected content, or focusing only on imitating local structures, which leads to the problem of local myopia. To overcome these limitations, we propose a dual-constraint stealthy node injection framework, called Joint Alignment of Nodal and Universal Structures (JANUS). At the local level, we introduce a local feature manifold alignment strategy to achieve geometric consistency in the feature space. At the global level, we incorporate structured latent variables and maximize the mutual information with the generated structures, ensuring the injected structures are consistent with the semantic patterns of the original graph. We model the injection attack as a sequential decision process, which is optimized by a reinforcement learning agent. Experiments on multiple standard datasets demonstrate that the JANUS framework significantly outperforms existing methods in terms of both attack effectiveness and stealthiness.
【2】On the Out-of-Distribution Backdoor Attack for Federated Learning
标题：联邦学习的分发外后门攻击
链接：https://arxiv.org/abs/2509.13219
备注：To appear at MobiHoc 2025
摘要：联邦学习（FL）中的传统后门攻击在受限的攻击场景中运行，因为它们依赖于可见的触发器，并且需要对目标对象进行物理修改，这限制了它们的实用性。为了解决这个问题，我们介绍了一种新的后门攻击原型FL称为外的分布（OOD）后门攻击（$\mathtt{OBA}$），它使用OOD数据作为中毒的样本和触发器同时。我们的方法显着扩大了FL中后门攻击场景的范围。为了提高$\marttt {OBA}$的隐蔽性，我们提出了$\marttt {SoDa}$，它在本地训练期间正则化恶意本地模型的大小和方向，使它们与良性版本紧密对齐以逃避检测。实证结果表明，$\mathtt{OBA}$有效地规避了最先进的防御，同时保持高精度的主要任务。 为了解决FL系统中的这个安全漏洞，我们引入了$\marttt {BNGuard}$，一种针对$\marttt {SoDa}$定制的新的服务器端防御方法。$\mathtt{BNGuard}$利用了OOD数据导致批处理规范化层的运行统计数据出现显著偏差的观察结果。这使得$\marttt {BNGuard}$能够识别恶意模型更新并将其从聚合中排除，从而增强FL的后门鲁棒性。在各种设置下的大量实验表明了$\marttt {BNGuard}$在防御$\marttt {SoDa}$方面的有效性。该代码可在https://github.com/JiiahaoXU/SoDa-BNGuard上获得。
摘要：Traditional backdoor attacks in federated learning (FL) operate within constrained attack scenarios, as they depend on visible triggers and require physical modifications to the target object, which limits their practicality. To address this limitation, we introduce a novel backdoor attack prototype for FL called the out-of-distribution (OOD) backdoor attack ($\mathtt{OBA}$), which uses OOD data as both poisoned samples and triggers simultaneously. Our approach significantly broadens the scope of backdoor attack scenarios in FL. To improve the stealthiness of $\mathtt{OBA}$, we propose $\mathtt{SoDa}$, which regularizes both the magnitude and direction of malicious local models during local training, aligning them closely with their benign versions to evade detection. Empirical results demonstrate that $\mathtt{OBA}$ effectively circumvents state-of-the-art defenses while maintaining high accuracy on the main task. To address this security vulnerability in the FL system, we introduce $\mathtt{BNGuard}$, a new server-side defense method tailored against $\mathtt{SoDa}$. $\mathtt{BNGuard}$ leverages the observation that OOD data causes significant deviations in the running statistics of batch normalization layers. This allows $\mathtt{BNGuard}$ to identify malicious model updates and exclude them from aggregation, thereby enhancing the backdoor robustness of FL. Extensive experiments across various settings show the effectiveness of $\mathtt{BNGuard}$ on defending against $\mathtt{SoDa}$. The code is available at https://github.com/JiiahaoXU/SoDa-BNGuard.
【3】BAPFL: Exploring Backdoor Attacks Against Prototype-based Federated Learning
标题：BAPFL：探索针对基于原型的联邦学习的后门攻击
链接：https://arxiv.org/abs/2509.12964
摘要：基于原型的联邦学习（Prototype-based Federated Learning，PFL）是解决联邦学习中数据异构性问题的一种很有前途的方法，它利用平均特征向量作为原型来提高模型的泛化能力。然而，它对后门攻击的鲁棒性在很大程度上仍未得到探索。在本文中，我们确定，PFL是固有的抵抗现有的后门攻击，由于其独特的原型学习机制和本地数据的异构性。为了进一步探索PFL的安全性，我们提出了BAPFL，这是第一个专门为PFL框架设计的后门攻击方法。BAPFL将原型中毒策略与触发优化机制相结合。原型中毒策略操纵全局原型的轨迹，以误导良性客户端的原型训练，将其本地干净样本的原型从嵌入式样本的原型中推出来。同时，触发器优化机制为每个潜在的目标标签学习一个唯一的、隐蔽的触发器，并引导嵌入触发器的样本原型与目标标签的全局原型紧密对齐。在多个数据集和PFL变体上的实验结果表明，与传统后门攻击相比，BAPFL在保持主任务准确性的同时，攻击成功率提高了35%~ 75%.这些结果突出了BAPFL在PFL中的有效性、隐蔽性和适应性。
摘要 ：Prototype-based federated learning (PFL) has emerged as a promising paradigm to address data heterogeneity problems in federated learning, as it leverages mean feature vectors as prototypes to enhance model generalization. However, its robustness against backdoor attacks remains largely unexplored. In this paper, we identify that PFL is inherently resistant to existing backdoor attacks due to its unique prototype learning mechanism and local data heterogeneity. To further explore the security of PFL, we propose BAPFL, the first backdoor attack method specifically designed for PFL frameworks. BAPFL integrates a prototype poisoning strategy with a trigger optimization mechanism. The prototype poisoning strategy manipulates the trajectories of global prototypes to mislead the prototype training of benign clients, pushing their local prototypes of clean samples away from the prototypes of trigger-embedded samples. Meanwhile, the trigger optimization mechanism learns a unique and stealthy trigger for each potential target label, and guides the prototypes of trigger-embedded samples to align closely with the global prototype of the target label. Experimental results across multiple datasets and PFL variants demonstrate that BAPFL achieves a 35\%-75\% improvement in attack success rate compared to traditional backdoor attacks, while preserving main task accuracy. These results highlight the effectiveness, stealthiness, and adaptability of BAPFL in PFL.
【4】Sy-FAR: Symmetry-based Fair Adversarial Robustness
标题：Sy-FAR：基于对称性的公平对抗鲁棒性
链接：https://arxiv.org/abs/2509.12939
备注：20 pages, 11 figures
摘要：安全关键机器学习（ML）系统，如人脸识别系统，容易受到对抗性示例的影响，包括现实世界中的物理可实现攻击。已经提出了各种方法来提高ML的对抗鲁棒性;然而，它们通常会导致不公平的鲁棒性：从某些类或组进行攻击通常比从其他类或组进行攻击更容易。已经开发了几种技术来提高对抗鲁棒性，同时寻求类之间的完美公平性。然而，先前的工作集中在安全和公平不那么重要的环境中。我们的观点是，在现实的公平关键任务（如人脸识别）中实现完美的对等通常是不可行的-某些类别可能非常相似，导致它们之间的错误分类。相反，我们建议寻求对称性-即，从类$i$到$j$的攻击会和从类$j$到$i$的攻击一样成功--更容易处理。直观地说，对称性是一种理想的，因为类相似性在大多数领域中是一种对称关系。此外，正如我们从理论上证明的那样，个体之间的对称性会导致任何一组子群体之间的对称性，这与其他公平概念相反，在这些公平概念中，群体公平往往是难以捉摸的。我们开发了Sy-FAR，这是一种鼓励对称性同时优化对抗鲁棒性的技术，并使用五个数据集，三个模型架构对其进行了广泛的评估，包括针对有针对性和无针对性的现实攻击。结果表明，与最先进的方法相比，Sy-FAR显着提高了公平对抗鲁棒性。此外，我们发现Sy-FAR在运行中更快，更一致。值得注意的是，Sy-FAR还改善了我们在这项工作中发现的另一种类型的不公平性--在引入对称性后，对抗性示例可能被分类的目标类变得不那么脆弱。
摘要：Security-critical machine-learning (ML) systems, such as face-recognition systems, are susceptible to adversarial examples, including real-world physically realizable attacks. Various means to boost ML's adversarial robustness have been proposed; however, they typically induce unfair robustness: It is often easier to attack from certain classes or groups than from others. Several techniques have been developed to improve adversarial robustness while seeking perfect fairness between classes. Yet, prior work has focused on settings where security and fairness are less critical. Our insight is that achieving perfect parity in realistic fairness-critical tasks, such as face recognition, is often infeasible -- some classes may be highly similar, leading to more misclassifications between them. Instead, we suggest that seeking symmetry -- i.e., attacks from class $i$ to $j$ would be as successful as from $j$ to $i$ -- is more tractable. Intuitively, symmetry is a desirable because class resemblance is a symmetric relation in most domains. Additionally, as we prove theoretically, symmetry between individuals induces symmetry between any set of sub-groups, in contrast to other fairness notions where group-fairness is often elusive. We develop Sy-FAR, a technique to encourage symmetry while also optimizing adversarial robustness and extensively evaluate it using five datasets, with three model architectures, including against targeted and untargeted realistic attacks. The results show Sy-FAR significantly improves fair adversarial robustness compared to state-of-the-art methods. Moreover, we find that Sy-FAR is faster and more consistent across runs. Notably, Sy-FAR also ameliorates another type of unfairness we discover in this work -- target classes that adversarial examples are likely to be classified into become significantly less vulnerable after inducing symmetry.
【5】Diffusion-Based Generation and Imputation of Driving Scenarios from Limited Vehicle CAN Data
标题：从有限的车辆CAN数据中基于扩散的驾驶场景生成和估算
链接：https://arxiv.org/abs/2509.12375
备注：Preprint, Paper has been accepted at ITSC 2025
摘要：在包含损坏样本的小型时间序列数据集上训练深度学习方法是一项挑战。扩散模型已被证明是有效的，以产生现实的和合成的数据，并通过插补纠正损坏的样本。在这种情况下，本文的重点是生成合成的汽车时间序列数据的真实样本。我们表明，去噪扩散概率模型（DDPM）可以有效地解决这一任务，将它们应用于具有长期数据和有限数量样本的具有挑战性的车辆CAN数据集。因此，我们提出了一种混合生成的方法，结合自回归和非自回归技术。我们评估我们的方法与两个最近提出的DDPM架构的时间序列生成，我们提出了一些改进。为了评估生成的样本，我们提出了三个指标，量化物理正确性和测试跟踪的坚持。我们最好的模型在物理正确性方面甚至优于训练数据，同时显示出合理的驾驶行为。最后，我们使用我们最好的模型成功地估算了训练数据中物理上不可信的区域，从而提高了数据质量。
摘要：Training deep learning methods on small time series datasets that also include corrupted samples is challenging. Diffusion models have shown to be effective to generate realistic and synthetic data, and correct corrupted samples through imputation. In this context, this paper focuses on generating synthetic yet realistic samples of automotive time series data. We show that denoising diffusion probabilistic models (DDPMs) can effectively solve this task by applying them to a challenging vehicle CAN-dataset with long-term data and a limited number of samples. Therefore, we propose a hybrid generative approach that combines autoregressive and non-autoregressive techniques. We evaluate our approach with two recently proposed DDPM architectures for time series generation, for which we propose several improvements. To evaluate the generated samples, we propose three metrics that quantify physical correctness and test track adherence. Our best model is able to outperform even the training data in terms of physical correctness, while showing plausible driving behavior. Finally, we use our best model to successfully impute physically implausible regions in the training data, thereby improving the data quality.
【6】Prediction of Stocks Index Price using Quantum GANs
标题：利用Quantum GAN预测股指价格
链接：https://arxiv.org/abs/2509.12286
摘要：本文研究了量子生成对抗网络（QGANs）在股票价格预测中的应用。金融市场本质上是复杂的，其特点是高波动性和复杂的模式，传统模型往往无法捕捉。QGAN利用量子计算的力量，通过将生成模型的优势与量子机器学习技术相结合，提供了一种新的方法。我们实现了一个专门用于股票价格预测的QGAN模型，并使用历史股票市场数据评估其性能。我们的研究结果表明，QGAN可以生成与实际市场行为非常相似的合成数据，从而提高预测准确性。实验使用股票指数价格数据和AWS Braket SV1模拟器来训练QGAN电路。量子增强模型在收敛速度和预测精度方面优于经典的长短期记忆（LSTM）和GAN模型。这项研究是将量子计算整合到金融预测中的关键一步，与传统方法相比，它在速度和精度方面具有潜在优势。这些发现对交易员、金融分析师和寻求先进市场分析工具的研究人员具有重要意义。
摘要 ：This paper investigates the application of Quantum Generative Adversarial Networks (QGANs) for stock price prediction. Financial markets are inherently complex, marked by high volatility and intricate patterns that traditional models often fail to capture. QGANs, leveraging the power of quantum computing, offer a novel approach by combining the strengths of generative models with quantum machine learning techniques. We implement a QGAN model tailored for stock price prediction and evaluate its performance using historical stock market data. Our results demonstrate that QGANs can generate synthetic data closely resembling actual market behavior, leading to enhanced prediction accuracy. The experiment was conducted using the Stocks index price data and the AWS Braket SV1 simulator for training the QGAN circuits. The quantum-enhanced model outperforms classical Long Short-Term Memory (LSTM) and GAN models in terms of convergence speed and prediction accuracy. This research represents a key step toward integrating quantum computing in financial forecasting, offering potential advantages in speed and precision over traditional methods. The findings suggest important implications for traders, financial analysts, and researchers seeking advanced tools for market analysis.
【7】InJecteD: Analyzing Trajectories and Drift Dynamics in Denoising Diffusion Probabilistic Models for 2D Point Cloud Generation
标题：InJecteD：分析2D点云生成去噪扩散概率模型中的轨迹和漂移动态
链接：https://arxiv.org/abs/2509.12239
摘要：这项工作介绍了InJecteD，一个框架，用于解释去噪扩散概率模型（DDPMs）通过分析样本轨迹在去噪过程中的2D点云生成。我们将此框架应用于Datasaurus Dozen bullseye，dino和circle的三个数据集，使用具有可定制输入和时间嵌入的简化DDPM架构。我们的方法量化的轨迹属性，包括位移，速度，聚类和漂移场动态，使用统计指标，如Wasserstein距离和余弦相似性。通过增强模型的透明度，InJecteD通过使从业者能够调试和改进生成模型来支持人类AI协作。实验揭示了不同的去噪阶段：初始噪声探索，快速形状形成和最终细化，以特定于小行星的行为为例，牛眼同心收敛与恐龙复杂轮廓形成。我们评估了四种模型配置，不同的嵌入和噪声时间表，表明基于傅立叶的嵌入提高了轨迹稳定性和重建质量
摘要：This work introduces InJecteD, a framework for interpreting Denoising Diffusion Probabilistic Models (DDPMs) by analyzing sample trajectories during the denoising process of 2D point cloud generation. We apply this framework to three datasets from the Datasaurus Dozen bullseye, dino, and circle using a simplified DDPM architecture with customizable input and time embeddings. Our approach quantifies trajectory properties, including displacement, velocity, clustering, and drift field dynamics, using statistical metrics such as Wasserstein distance and cosine similarity. By enhancing model transparency, InJecteD supports human AI collaboration by enabling practitioners to debug and refine generative models. Experiments reveal distinct denoising phases: initial noise exploration, rapid shape formation, and final refinement, with dataset-specific behaviors example, bullseyes concentric convergence vs. dinos complex contour formation. We evaluate four model configurations, varying embeddings and noise schedules, demonstrating that Fourier based embeddings improve trajectory stability and reconstruction quality
半/弱/无/有监督|不确定性|主动学习(10篇)
【1】Post-Hoc Split-Point Self-Consistency Verification for Efficient, Unified Quantification of Aleatoric and Epistemic Uncertainty in Deep Learning
标题：事后分点自一致性验证，用于深度学习中的感性和认识不确定性的高效、统一量化
链接：https://arxiv.org/abs/2509.13262
备注：32 pages, 15 figures and 16 tables. Technical Report submitted to a journal for publication
摘要：不确定性量化（UQ）对于值得信赖的深度学习至关重要，但现有的方法要么是计算密集型的，如贝叶斯或集成方法，要么只提供部分特定于任务的估计，如单次向前传递技术。在本文中，我们提出了一个事后单前向传递框架，该框架可以在不修改或重新训练预训练模型的情况下，联合捕获任意和认知的不确定性。我们的方法应用分裂点分析（SPA）将预测残差分解为上下子集，计算每一侧的平均绝对残差（MAR）。我们证明了，在理想条件下，总MAR等于调和平均子集MAR的偏差定义了一个新的\emdash {自我一致性离散分数}（SDS）的细粒度的认知估计跨越回归和分类。对于回归，侧特异性分位数回归产生具有改进的经验覆盖率的预测区间，其通过SDS进一步校准。对于分类，当校准数据可用时，我们应用基于SPA的校准标识来调整softmax输出，然后计算这些校准概率的预测熵。在不同的回归和分类基准上进行的大量实验表明，我们的框架匹配或超过了几种最先进的UQ方法，同时产生的开销最小。 我们的源代码可在https://github.com/zzz0527/SPC-UQ上获得。
摘要：Uncertainty quantification (UQ) is vital for trustworthy deep learning, yet existing methods are either computationally intensive, such as Bayesian or ensemble methods, or provide only partial, task-specific estimates, such as single-forward-pass techniques. In this paper, we propose a post-hoc single-forward-pass framework that jointly captures aleatoric and epistemic uncertainty without modifying or retraining pretrained models. Our method applies \emph{Split-Point Analysis} (SPA) to decompose predictive residuals into upper and lower subsets, computing \emph{Mean Absolute Residuals} (MARs) on each side. We prove that, under ideal conditions, the total MAR equals the harmonic mean of subset MARs; deviations define a novel \emph{Self-consistency Discrepancy Score} (SDS) for fine-grained epistemic estimation across regression and classification. For regression, side-specific quantile regression yields prediction intervals with improved empirical coverage, which are further calibrated via SDS. For classification, when calibration data are available, we apply SPA-based calibration identities to adjust the softmax outputs and then compute predictive entropy on these calibrated probabilities. Extensive experiments on diverse regression and classification benchmarks demonstrate that our framework matches or exceeds several state-of-the-art UQ methods while incurring minimal overhead. Our source code is available at https://github.com/zzz0527/SPC-UQ.
【2】Curriculum Multi-Task Self-Supervision Improves Lightweight Architectures for Onboard Satellite Hyperspectral Image Segmentation
标题：课程多任务自我监督改进星载卫星高光谱图像分割的轻量级架构
链接：https://arxiv.org/abs/2509.13229
摘要：高光谱成像（HSI）捕获每个像素数百个连续波段的详细光谱特征，对于土地覆盖分类，变化检测和环境监测等遥感应用是不可或缺的。由于HSI数据的高维度和基于卫星的系统中的数据传输速率缓慢，需要紧凑和高效的模型来支持机载处理，并最大限度地减少冗余或低值数据的传输，例如云覆盖区域。为此，我们引入了一个新的课程多任务自监督学习（CMTSSL）框架，设计用于HSI分析的轻量级架构。CMTSSL将掩蔽图像建模与解耦的空间和光谱拼图解决方案相结合，并在课程学习策略的指导下，在自我监督过程中逐步增加数据的复杂性。这使得编码器能够联合捕获细粒度的频谱连续性、空间结构和全局语义特征。与先前的双任务SSL方法不同，CMTSSL在统一且计算效率高的设计中同时解决了空间和光谱推理，特别适合于训练用于机载卫星部署的轻量级模型。我们在四个公共基准数据集上验证了我们的方法，证明了下游分割任务的一致收益，使用的架构比一些最先进的模型轻16，000倍以上。这些结果强调了CMTSSL在现实世界HSI应用程序的轻量级架构的可概括表示学习中的潜力。我们的代码可在https://github.com/hugocarlesso/CMTSSL上公开获取。
摘要 ：Hyperspectral imaging (HSI) captures detailed spectral signatures across hundreds of contiguous bands per pixel, being indispensable for remote sensing applications such as land-cover classification, change detection, and environmental monitoring. Due to the high dimensionality of HSI data and the slow rate of data transfer in satellite-based systems, compact and efficient models are required to support onboard processing and minimize the transmission of redundant or low-value data, e.g. cloud-covered areas. To this end, we introduce a novel curriculum multi-task self-supervised learning (CMTSSL) framework designed for lightweight architectures for HSI analysis. CMTSSL integrates masked image modeling with decoupled spatial and spectral jigsaw puzzle solving, guided by a curriculum learning strategy that progressively increases data complexity during self-supervision. This enables the encoder to jointly capture fine-grained spectral continuity, spatial structure, and global semantic features. Unlike prior dual-task SSL methods, CMTSSL simultaneously addresses spatial and spectral reasoning within a unified and computationally efficient design, being particularly suitable for training lightweight models for onboard satellite deployment. We validate our approach on four public benchmark datasets, demonstrating consistent gains in downstream segmentation tasks, using architectures that are over 16,000x lighter than some state-of-the-art models. These results highlight the potential of CMTSSL in generalizable representation learning with lightweight architectures for real-world HSI applications. Our code is publicly available at https://github.com/hugocarlesso/CMTSSL.
【3】TRUST-FS: Tensorized Reliable Unsupervised Multi-View Feature Selection for Incomplete Data
标题：TRUST-FS：针对不完整数据的张量化可靠无监督多视图特征选择
链接：https://arxiv.org/abs/2509.13192
摘要：多视图无监督特征选择（MUFS）是从多视图未标记数据中选择信息特征的一种方法，近年来引起了越来越多的研究兴趣。虽然人们已经在MUFS上投入了大量的努力，但仍然存在一些挑战：1）现有的用于不完整多视图数据的方法仅限于处理缺失视图，并且无法解决更一般的缺失变量场景，其中某些特征在某些视图中具有缺失值; 2）大多数方法通过首先估算缺失值然后执行特征选择来解决不完整数据，独立处理这两个过程，忽略了它们的相互作用; 3）丢失数据可能导致不准确的相似性图，这降低了特征选择的性能。为了解决这个难题，我们提出了一种新的MUFS方法不完整的多视图数据与丢失的变量，称为张量可靠的无监督多视图特征选择（TRUST-FS）。TRUST-FS引入了一种新的自适应加权CP分解，可以在统一的张量分解框架内同时执行特征选择、缺失变量插补和视图权重学习。通过利用主观逻辑来获取可信的跨视图相似性信息，TRUST-FS有助于学习可靠的相似性图，从而指导特征选择和填补。综合实验结果表明，我们的方法比国家的最先进的方法的有效性和优越性。
摘要：Multi-view unsupervised feature selection (MUFS), which selects informative features from multi-view unlabeled data, has attracted increasing research interest in recent years. Although great efforts have been devoted to MUFS, several challenges remain: 1) existing methods for incomplete multi-view data are limited to handling missing views and are unable to address the more general scenario of missing variables, where some features have missing values in certain views; 2) most methods address incomplete data by first imputing missing values and then performing feature selection, treating these two processes independently and overlooking their interactions; 3) missing data can result in an inaccurate similarity graph, which reduces the performance of feature selection. To solve this dilemma, we propose a novel MUFS method for incomplete multi-view data with missing variables, termed Tensorized Reliable UnSupervised mulTi-view Feature Selection (TRUST-FS). TRUST-FS introduces a new adaptive-weighted CP decomposition that simultaneously performs feature selection, missing-variable imputation, and view weight learning within a unified tensor factorization framework. By utilizing Subjective Logic to acquire trustworthy cross-view similarity information, TRUST-FS facilitates learning a reliable similarity graph, which subsequently guides feature selection and imputation. Comprehensive experimental results demonstrate the effectiveness and superiority of our method over state-of-the-art methods.
【4】Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropy
标题：元学习已经过时了吗？重新思考具有有限熵的无监督Few-Shot分类
链接：https://arxiv.org/abs/2509.13185
备注：Accepted by ICCV 2025
摘要：元学习是处理Few-Shot任务的强大范例。然而，最近的研究表明，在Few-Shot分类任务中，使用整类训练策略训练的模型可以达到与使用元学习训练的模型相当的性能。为了证明元学习的价值，我们建立了一个熵限制的监督设置，以进行公平的比较。通过理论分析和实验验证，我们建立了元学习有一个更严格的泛化界相比，整个类的训练。我们发现，元学习在有限熵的情况下更有效，对标签噪声和异构任务更鲁棒，因此非常适合无监督任务。基于这些见解，我们提出了MINO，一个旨在提高无监督性能的元学习框架。MINO利用自适应聚类算法DBSCAN，具有用于无监督任务构建的动态头部和用于对标签噪声鲁棒性的基于稳定性的元缩放器。大量的实验证实了它在多个无监督的Few-Shot和zero-shot任务中的有效性。
摘要：Meta-learning is a powerful paradigm for tackling few-shot tasks. However, recent studies indicate that models trained with the whole-class training strategy can achieve comparable performance to those trained with meta-learning in few-shot classification tasks. To demonstrate the value of meta-learning, we establish an entropy-limited supervised setting for fair comparisons. Through both theoretical analysis and experimental validation, we establish that meta-learning has a tighter generalization bound compared to whole-class training. We unravel that meta-learning is more efficient with limited entropy and is more robust to label noise and heterogeneous tasks, making it well-suited for unsupervised tasks. Based on these insights, We propose MINO, a meta-learning framework designed to enhance unsupervised performance. MINO utilizes the adaptive clustering algorithm DBSCAN with a dynamic head for unsupervised task construction and a stability-based meta-scaler for robustness against label noise. Extensive experiments confirm its effectiveness in multiple unsupervised few-shot and zero-shot tasks.
【5】Deep Generative and Discriminative Digital Twin endowed with Variational Autoencoder for Unsupervised Predictive Thermal Condition Monitoring of Physical Robots in Industry 6.0 and Society 6.0
标题：深度生成和区分性数字双胞胎配备变分自动编码器，用于工业6.0和社会6.0中物理机器人的无监督预测热状况监控
链接：https://arxiv.org/abs/2509.12740
备注：$©$ 2025 the authors. This work has been accepted to the to the 10th IFAC Symposium on Mechatronic Systems & 14th IFAC Symposium on Robotics July 15-18, 2025 || Paris, France for publication under a Creative Commons Licence CC-BY-NC-ND
摘要：机器人在工业4.0中被广泛用于提高运营效率，并为工业5.0中的劳动力提供共生和可持续的帮助。由于抗脆弱制造和以人为中心的社会任务需要弹性，鲁棒性和福祉，因此对电机过热引起的热饱和和烧伤的自主预测和适应对于人类安全和机器人可用性至关重要。因此，机器人有望自我维持其性能并提供用户体验，此外还将其能力提前传达给其他代理，以确保完全自动化的热可行任务，并在没有人为干预的情况下延长其寿命。然而，传统的机器人关机，当面临迫在眉睫的热饱和，抑制了工厂的生产力和社会的舒适度，而冷却策略是很难实施后，机器人收购。在这项工作中，智能数字双胞胎被赋予了生成AI，即，可变自动编码器被用来管理热异常并生成不关键的机器人状态。热困难的概念来自于变分自编码器的重构误差。机器人可以使用该分数来预测、预测和共享所需运动轮廓的热可行性，以满足工业6.0和社会6.0中新兴应用的要求。
摘要：Robots are unrelentingly used to achieve operational efficiency in Industry 4.0 along with symbiotic and sustainable assistance for the work-force in Industry 5.0. As resilience, robustness, and well-being are required in anti-fragile manufacturing and human-centric societal tasks, an autonomous anticipation and adaption to thermal saturation and burns due to motors overheating become instrumental for human safety and robot availability. Robots are thereby expected to self-sustain their performance and deliver user experience, in addition to communicating their capability to other agents in advance to ensure fully automated thermally feasible tasks, and prolong their lifetime without human intervention. However, the traditional robot shutdown, when facing an imminent thermal saturation, inhibits productivity in factories and comfort in the society, while cooling strategies are hard to implement after the robot acquisition. In this work, smart digital twins endowed with generative AI, i.e., variational autoencoders, are leveraged to manage thermally anomalous and generate uncritical robot states. The notion of thermal difficulty is derived from the reconstruction error of variational autoencoders. A robot can use this score to predict, anticipate, and share the thermal feasibility of desired motion profiles to meet requirements from emerging applications in Industry 6.0 and Society 6.0.
【6】Bayesian Parametric Matrix Models: Principled Uncertainty Quantification for Spectral Learning
标题：Bayesian参数矩阵模型：谱学习的原则不确定性量化
链接：https://arxiv.org/abs/2509.12406
摘要：科学机器学习越来越多地使用光谱方法来理解物理系统。目前的谱学习方法只提供点估计，没有不确定性量化，限制了它们在预测置信度至关重要的安全关键应用中的使用。参数矩阵模型已经成为科学机器学习的强大工具，通过学习控制方程实现卓越的性能。然而，它们的确定性限制了在不确定性量化应用中的部署。我们介绍贝叶斯参数矩阵模型（B-PMM），一个原则性的框架，扩展PMM提供不确定性估计，同时保持其频谱结构和计算效率。B-PMM解决了量化矩阵特征值问题中的不确定性的根本挑战，其中标准贝叶斯方法由于谱分解的几何约束而失败。理论贡献包括：（一）自适应谱分解与正则化矩阵扰动界特征的特征值不确定性传播，（二）结构化变分推理算法使用流形感知矩阵变量高斯后验，尊重埃尔米特约束，和（iii）有限样本校准保证明确依赖于频谱间隙和问题条件。实验验证矩阵尺寸从5x 5到500 x500，具有完美的收敛速度，表明B-PMM实现了出色的不确定度校准（ECE < 0.05），同时保持了良好的缩放。该框架表现出优雅的光谱病态下的退化，并提供可靠的不确定性估计，即使在近退化制度。该框架支持不确定性关键域中的鲁棒谱学习，并为更广泛的贝叶斯谱机器学习奠定了基础。
摘要：Scientific machine learning increasingly uses spectral methods to understand physical systems. Current spectral learning approaches provide only point estimates without uncertainty quantification, limiting their use in safety-critical applications where prediction confidence is essential. Parametric matrix models have emerged as powerful tools for scientific machine learning, achieving exceptional performance by learning governing equations. However, their deterministic nature limits deployment in uncertainty quantification applications. We introduce Bayesian parametric matrix models (B-PMMs), a principled framework that extends PMMs to provide uncertainty estimates while preserving their spectral structure and computational efficiency. B-PMM addresses the fundamental challenge of quantifying uncertainty in matrix eigenvalue problems where standard Bayesian methods fail due to the geometric constraints of spectral decomposition. The theoretical contributions include: (i) adaptive spectral decomposition with regularized matrix perturbation bounds that characterize eigenvalue uncertainty propagation, (ii) structured variational inference algorithms using manifold-aware matrix-variate Gaussian posteriors that respect Hermitian constraints, and (iii) finite-sample calibration guarantees with explicit dependence on spectral gaps and problem conditioning. Experimental validation across matrix dimensions from 5x5 to 500x500 with perfect convergence rates demonstrates that B-PMMs achieve exceptional uncertainty calibration (ECE < 0.05) while maintaining favorable scaling. The framework exhibits graceful degradation under spectral ill-conditioning and provides reliable uncertainty estimates even in near-degenerate regimes. The proposed framework supports robust spectral learning in uncertainty-critical domains and lays the groundwork for broader Bayesian spectral machine learning.
【7】Explainable Unsupervised Multi-Anomaly Detection and Temporal Localization in Nuclear Times Series Data with a Dual Attention-Based Autoencoder
标题：使用基于双注意的自动编码器在核时间序列数据中进行可解释的无监督多异常检测和时间定位
链接：https://arxiv.org/abs/2509.12372
摘要：核工业正在朝着更多新的反应堆设计方向发展，预计下一代反应堆的规模和功率输出将更小。这些系统有可能以多元时间序列数据的形式产生大量信息，可用于增强实时监测和控制。在这种情况下，远程自主或半自主控制系统的反应堆操作的发展已经获得了显着的兴趣。实现这种系统的关键第一步是能够检测和定位反应堆系统内的异常的准确诊断模块。最近的研究已经提出了各种ML和DL方法在核领域的异常检测。尽管取得了可喜的成果，但关键挑战仍然存在，包括有限的无法解释性，缺乏对真实世界数据的访问，以及异常事件的稀缺，这阻碍了基准测试和表征。大多数现有的研究将这些方法视为黑箱，而最近的工作强调了在安全关键领域中需要更大的ML/DL输出的可解释性。在这里，我们提出了一种基于LSTM自动编码器的无监督方法，该方法具有双重注意力机制，用于表征真实世界反应堆辐射区域监测系统中的异常事件。该框架不仅包括检测，但也定位的事件，并使用真实世界的数据集的复杂性不断增加的PUR-1研究反应堆进行了评估。注意力机制在特征和时间维度上操作，其中特征注意力将权重分配给表现出异常模式的辐射传感器，而时间注意力突出显示发生不规则的特定时间步，从而实现定位。通过结合结果，该框架可以在单个统一网络中识别受影响的传感器和每个异常的持续时间。
摘要：The nuclear industry is advancing toward more new reactor designs, with next-generation reactors expected to be smaller in scale and power output. These systems have the potential to produce large volumes of information in the form of multivariate time-series data, which could be used for enhanced real-time monitoring and control. In this context, the development of remote autonomous or semi-autonomous control systems for reactor operation has gained significant interest. A critical first step toward such systems is an accurate diagnostics module capable of detecting and localizing anomalies within the reactor system. Recent studies have proposed various ML and DL approaches for anomaly detection in the nuclear domain. Despite promising results, key challenges remain, including limited to no explainability, lack of access to real-world data, and scarcity of abnormal events, which impedes benchmarking and characterization. Most existing studies treat these methods as black boxes, while recent work highlights the need for greater interpretability of ML/DL outputs in safety-critical domains. Here, we propose an unsupervised methodology based on an LSTM autoencoder with a dual attention mechanism for characterization of abnormal events in a real-world reactor radiation area monitoring system. The framework includes not only detection but also localization of the event and was evaluated using real-world datasets of increasing complexity from the PUR-1 research reactor. The attention mechanisms operate in both the feature and temporal dimensions, where the feature attention assigns weights to radiation sensors exhibiting abnormal patterns, while time attention highlights the specific timesteps where irregularities occur, thus enabling localization. By combining the results, the framework can identify both the affected sensors and the duration of each anomaly within a single unified network.
【8】Uncertainty-Aware Hourly Air Temperature Mapping at 2 km Resolution via Physics-Guided Deep Learning
标题：通过物理引导的深度学习以2公里分辨率绘制具有不确定性的小时气温地图
链接：https://arxiv.org/abs/2509.12329
摘要：近地表气温是地球表面的一个重要物理特性。虽然气象站提供连续监测，卫星提供广泛的空间覆盖，但没有单一的数据源以时空方式提供无缝数据。在这里，我们提出了一种数据驱动的、物理学指导的深度学习方法，以2 km的分辨率在美国本土生成每小时的气温数据。这种方法被称为放大器空气变压器，首先重建被云层遮挡的GOES-16表面温度数据。它通过一个用年温度周期编码的神经网络来实现这一点，其中包含一个线性项，以更精细的尺度放大ERA 5温度值，并通过卷积层来捕获时空变化。然后，另一个神经网络通过利用其与关键地球表面特性的潜在关系将重建的表面温度转换为空气温度。该方法通过深度集成学习的预测不确定性估计进一步增强，以提高可靠性。所提出的方法是在美国邻近地区（2018-2024年）气象站的777亿表面温度像素和1.55亿空气温度记录上构建和测试的，在基于站点的验证中实现了1.93 C的每小时空气温度映射精度。该方法简化了地面温度重建和气温预测，它可以扩展到其他卫星源的无缝空气温度监测在高时空分辨率。本研究生成的数据可在https：//doi.org/10.5281/zenodo.15252812下载，项目网页可在https://skrisliu.com/HourlyAirTemp2kmUSA/找到。
摘要：Near-surface air temperature is a key physical property of the Earth's surface. Although weather stations offer continuous monitoring and satellites provide broad spatial coverage, no single data source offers seamless data in a spatiotemporal fashion. Here, we propose a data-driven, physics-guided deep learning approach to generate hourly air temperature data at 2 km resolution over the contiguous United States. The approach, called Amplifier Air-Transformer, first reconstructs GOES-16 surface temperature data obscured by clouds. It does so through a neural network encoded with the annual temperature cycle, incorporating a linear term to amplify ERA5 temperature values at finer scales and convolutional layers to capture spatiotemporal variations. Then, another neural network transforms the reconstructed surface temperature into air temperature by leveraging its latent relationship with key Earth surface properties. The approach is further enhanced with predictive uncertainty estimation through deep ensemble learning to improve reliability. The proposed approach is built and tested on 77.7 billion surface temperature pixels and 155 million air temperature records from weather stations across the contiguous United States (2018-2024), achieving hourly air temperature mapping accuracy of 1.93 C in station-based validation. The proposed approach streamlines surface temperature reconstruction and air temperature prediction, and it can be extended to other satellite sources for seamless air temperature monitoring at high spatiotemporal resolution. The generated data of this study can be downloaded at https://doi.org/10.5281/zenodo.15252812, and the project webpage can be found at https://skrisliu.com/HourlyAirTemp2kmUSA/.
【9】SURGIN: SURrogate-guided Generative INversion for subsurface multiphase flow with quantified uncertainty
标题：SSE：具有量化不确定性的地下多相流的代理引导生成逆转换
链接：https://arxiv.org/abs/2509.13189
摘要：我们提出了一个直接的反演建模方法名为SURGIN，一个SURrogate引导生成反演框架尾随地下多相流数据同化。与现有的反演方法需要适应每个新的观测配置不同，SURGIN具有zero-shot条件生成能力，能够实时同化看不见的监测数据，而无需进行特定任务的再训练。具体而言，SURGIN协同集成了U-Net增强傅立叶神经运算符（U-FNO）替代与基于分数的生成模型（SGM），将条件生成框架为贝叶斯视角中的替代预测指导过程。与直接学习地质参数的条件生成不同，首先以自监督的方式预训练无条件SGM以捕获地质先验，然后通过利用可微U-FNO替代来执行后验采样，以实现以看不见的观测为条件的有效前向评估。大量的数值实验表明，SURGIN的能力，体面地推断异质地质领域和预测时空流动动力学与量化的不确定性在不同的测量设置。通过将生成学习与代理引导的贝叶斯推理相结合，SURGIN为参数函数空间中的逆建模和不确定性量化建立了一个新的范式。
摘要：We present a direct inverse modeling method named SURGIN, a SURrogate-guided Generative INversion framework tailed for subsurface multiphase flow data assimilation. Unlike existing inversion methods that require adaptation for each new observational configuration, SURGIN features a zero-shot conditional generation capability, enabling real-time assimilation of unseen monitoring data without task-specific retraining. Specifically, SURGIN synergistically integrates a U-Net enhanced Fourier Neural Operator (U-FNO) surrogate with a score-based generative model (SGM), framing the conditional generation as a surrogate prediction-guidance process in a Bayesian perspective. Instead of directly learning the conditional generation of geological parameters, an unconditional SGM is first pretrained in a self-supervised manner to capture the geological prior, after which posterior sampling is performed by leveraging a differentiable U-FNO surrogate to enable efficient forward evaluations conditioned on unseen observations. Extensive numerical experiments demonstrate SURGIN's capability to decently infer heterogeneous geological fields and predict spatiotemporal flow dynamics with quantified uncertainty across diverse measurement settings. By unifying generative learning with surrogate-guided Bayesian inference, SURGIN establishes a new paradigm for inverse modeling and uncertainty quantification in parametric functional spaces.
【10】MEGAN: Mixture of Experts for Robust Uncertainty Estimation in Endoscopy Videos
标题：MEGAN：内窥镜检查中稳健不确定性估计的专家混合视频
链接：https://arxiv.org/abs/2509.12772
备注：11 pages, 2 figures, 1 table, accepted at UNSURE, MICCAI
摘要：可靠的不确定性量化（UQ）在医疗AI中至关重要。证据深度学习（EDL）提供了一种计算效率高的方法来量化模型的不确定性以及预测，这与传统的方法（如Monte Carlo（MC）Dropout和Deep Ensembles（DE））不同。然而，所有这些方法通常依赖于单个专家的注释作为模型训练的基础事实，忽略了医疗保健中的评分者间差异。为了解决这个问题，我们提出了MEGAN，这是一个多专家门控网络，它通过使用不同的地面事实和建模策略训练的EDL模型，聚合来自多个AI专家的不确定性估计和预测。MEGAN的门控网络将每个EDL模型的预测和不确定性最佳地结合起来，提高了整体预测的置信度和校准。我们在内镜视频上广泛地对MEGAN进行基准测试，以估计溃疡性结肠炎（UC）疾病的严重程度，通过Mayo内镜子评分（MES）的视觉标记进行评估，其中评分者间的差异很普遍。在大规模前瞻性UC临床试验中，与现有方法相比，MEGAN实现了F1评分提高3.5%，预期校准误差（ECE）降低30.5%。此外，MEGAN促进了不确定性指导的样本分层，减少了注释负担，并可能提高UC试验的效率和一致性。
摘要：Reliable uncertainty quantification (UQ) is essential in medical AI. Evidential Deep Learning (EDL) offers a computationally efficient way to quantify model uncertainty alongside predictions, unlike traditional methods such as Monte Carlo (MC) Dropout and Deep Ensembles (DE). However, all these methods often rely on a single expert's annotations as ground truth for model training, overlooking the inter-rater variability in healthcare. To address this issue, we propose MEGAN, a Multi-Expert Gating Network that aggregates uncertainty estimates and predictions from multiple AI experts via EDL models trained with diverse ground truths and modeling strategies. MEGAN's gating network optimally combines predictions and uncertainties from each EDL model, enhancing overall prediction confidence and calibration. We extensively benchmark MEGAN on endoscopy videos for Ulcerative colitis (UC) disease severity estimation, assessed by visual labeling of Mayo Endoscopic Subscore (MES), where inter-rater variability is prevalent. In large-scale prospective UC clinical trial, MEGAN achieved a 3.5% improvement in F1-score and a 30.5% reduction in Expected Calibration Error (ECE) compared to existing methods. Furthermore, MEGAN facilitated uncertainty-guided sample stratification, reducing the annotation burden and potentially increasing efficiency and consistency in UC trials.
迁移|Zero/Few/One-Shot|自适应(4篇)
【1】Prediction and Causality of functional MRI and synthetic signal using a Zero-Shot Time-Series Foundation Model
标题：使用Zero-Shot时间序列基础模型预测功能性MRI和合成信号及其因果关系
链接：https://arxiv.org/abs/2509.12497
摘要：时间序列预测和因果发现是神经科学的核心，因为预测大脑活动和识别神经群体和电路之间的因果关系可以揭示认知和疾病的潜在机制。随着基础模型的兴起，一个悬而未决的问题是它们与传统的脑信号预测和因果关系分析方法相比如何，以及它们是否可以应用于zero-shot设置。在这项工作中，我们评估了一个基础模型对经典的方法推断方向的相互作用，从自发脑活动测量功能性磁共振成像（fMRI）在人类。传统的方法往往依赖于维纳-格兰杰因果关系。我们测试了基础模型在zero-shot和微调设置下的预测能力，并通过比较模型的Granger类估计值与标准Granger因果关系来评估因果关系。我们使用从地面实况因果模型生成的合成时间序列验证了该方法，包括逻辑映射耦合和Ornstein-Uhlenbeck过程。基础模型实现了竞争性zero-shot预测fMRI时间序列（对照组的平均绝对百分比误差为0.55，患者为0.27）。虽然标准的格兰杰因果关系并没有显示出明确的定量模型之间的差异，基础模型提供了一个更精确的检测因果关系的相互作用。 总体而言，这些研究结果表明，基础模型提供了多功能性，强大的zero-shot性能，并在时间序列数据的预测和因果发现的潜在效用。
摘要：Time-series forecasting and causal discovery are central in neuroscience, as predicting brain activity and identifying causal relationships between neural populations and circuits can shed light on the mechanisms underlying cognition and disease. With the rise of foundation models, an open question is how they compare to traditional methods for brain signal forecasting and causality analysis, and whether they can be applied in a zero-shot setting. In this work, we evaluate a foundation model against classical methods for inferring directional interactions from spontaneous brain activity measured with functional magnetic resonance imaging (fMRI) in humans. Traditional approaches often rely on Wiener-Granger causality. We tested the forecasting ability of the foundation model in both zero-shot and fine-tuned settings, and assessed causality by comparing Granger-like estimates from the model with standard Granger causality. We validated the approach using synthetic time series generated from ground-truth causal models, including logistic map coupling and Ornstein-Uhlenbeck processes. The foundation model achieved competitive zero-shot forecasting fMRI time series (mean absolute percentage error of 0.55 in controls and 0.27 in patients). Although standard Granger causality did not show clear quantitative differences between models, the foundation model provided a more precise detection of causal interactions. Overall, these findings suggest that foundation models offer versatility, strong zero-shot performance, and potential utility for forecasting and causal discovery in time-series data.
【2】Adaptive Spatial Goodness Encoding: Advancing and Scaling Forward-Forward Learning Without Backpropagation
标题：自适应空间优度编码：在没有反向传播的情况下推进和缩放前向学习
链接：https://arxiv.org/abs/2509.12394
摘要：Forward-Forward（FF）算法为反向传播（BP）算法提供了一种很有前途的替代方案。尽管最近基于FF的扩展取得了进步，这些扩展增强了原始算法并使其适应卷积神经网络（CNN），但它们通常具有有限的表示能力和对大规模数据集的可扩展性差，这主要是由于信道维度的爆炸。在这项工作中，我们提出了自适应空间优度编码（ASGE），这是一种为CNN量身定制的新的基于FF的训练框架。ASGE利用特征图来计算每层的空间感知良好度表示，从而实现逐层监督。至关重要的是，这种方法将分类复杂性从通道维度中分离出来，从而解决了通道爆炸的问题，并与其他无BP方法相比实现了具有竞争力的性能。ASGE在多个基准测试中优于所有其他基于FF的方法，在MNIST上的测试准确率为99.65%，在FashionMNIST上为93.41%，在CIFAR-10上为90.62%，在CIFAR-100上为65.42%。此外，我们首次成功地将基于FF的训练应用于ImageNet，Top-1和Top-5的准确率分别为26.21%和47.49%。通过完全消除BP并显著缩小与BP训练模型的性能差距，ASGE框架为可扩展的无BP CNN训练建立了可行的基础。
摘要：The Forward-Forward (FF) algorithm offers a promising al- ternative to backpropagation (BP). Despite advancements in recent FF-based extensions, which have enhanced the origi- nal algorithm and adapted it to convolutional neural networks (CNNs), they often suffer from limited representational ca- pacity and poor scalability to large-scale datasets, primarily due to exploding channel dimensionality. In this work, we propose adaptive spatial goodness encoding (ASGE), a new FF-based training framework tailored for CNNs. ASGE lever- ages feature maps to compute spatially-aware goodness rep- resentations at each layer, enabling layer-wise supervision. Crucially, this approach decouples classification complexity from channel dimensionality, thereby addressing the issue of channel explosion and achieving competitive performance compared to other BP-free methods. ASGE outperforms all other FF-based approaches across multiple benchmarks, delivering test accuracies of 99.65% on MNIST, 93.41% on FashionMNIST, 90.62% on CIFAR-10, and 65.42% on CIFAR-100. Moreover, we present the first successful ap- plication of FF-based training to ImageNet, with Top-1 and Top-5 accuracies of 26.21% and 47.49%. By entirely elimi- nating BP and significantly narrowing the performance gap with BP-trained models, the ASGE framework establishes a viable foundation toward scalable BP-free CNN training.
【3】Causal-Symbolic Meta-Learning (CSML): Inducing Causal World Models for Few-Shot Generalization
标题：因果符号元学习（CSML）：为Few-Shot概括引入因果世界模型
链接：https://arxiv.org/abs/2509.12387
备注：10 pages, 4 figures
摘要：现代深度学习模型在模式识别方面表现出色，但仍然受到依赖虚假相关性的根本限制，导致泛化能力差，需要大量数据集。我们认为，一个关键因素，为人类一样的智能，强大的，样本有效的学习源于因果机制的理解。在这项工作中，我们介绍了因果符号元学习（CSML），一种新的框架，学习推断任务分布的潜在因果结构。CSML包括三个关键模块：感知模块，将原始输入映射到解开的符号表示;可微因果归纳模块，发现管理这些符号的潜在因果图;以及基于图的推理模块，利用该图进行预测。通过在任务分布中元学习共享的因果世界模型，CSML可以快速适应新的任务，包括那些需要从少数例子中推理干预和反事实的任务。我们介绍了Causal World，一个新的基于物理的基准测试，旨在测试这些功能。我们的实验表明，CSML大大优于最先进的元学习和神经符号基线，特别是在需要真正的因果推理的任务。
摘要：Modern deep learning models excel at pattern recognition but remain fundamentally limited by their reliance on spurious correlations, leading to poor generalization and a demand for massive datasets. We argue that a key ingredient for human-like intelligence-robust, sample-efficient learning-stems from an understanding of causal mechanisms. In this work, we introduce Causal-Symbolic Meta-Learning (CSML), a novel framework that learns to infer the latent causal structure of a task distribution. CSML comprises three key modules: a perception module that maps raw inputs to disentangled symbolic representations; a differentiable causal induction module that discovers the underlying causal graph governing these symbols and a graph-based reasoning module that leverages this graph to make predictions. By meta-learning a shared causal world model across a distribution of tasks, CSML can rapidly adapt to novel tasks, including those requiring reasoning about interventions and counterfactuals, from only a handful of examples. We introduce CausalWorld, a new physics-based benchmark designed to test these capabilities. Our experiments show that CSML dramatically outperforms state-of-the-art meta-learning and neuro-symbolic baselines, particularly on tasks demanding true causal inference.
【4】Learning to Route: Per-Sample Adaptive Routing for Multimodal Multitask Prediction
标题：学习路由：用于多模式多任务预测的按样本自适应路由
链接：https://arxiv.org/abs/2509.12227
摘要：我们提出了一个统一的框架，自适应路由多任务，多模态预测设置数据的异质性和任务之间的相互作用不同的样本。在心理治疗中，结构化的评估和非结构化的临床医生笔记共存的部分缺失的数据和相关的结果的应用程序的动机，我们介绍了一个路由为基础的架构，动态选择模态处理途径和任务共享策略的基础上，每个样本。我们的模型定义了多个模态路径，包括文本和数字特征的原始和融合表示，并学习通过信息量最大的专家组合来路由每个输入。任务特定的预测由共享或独立的头根据路由决策产生，整个系统是端到端训练的。我们评估了合成数据和现实世界的心理治疗笔记预测抑郁和焦虑的结果模型。我们的实验表明，我们的方法始终优于固定的多任务或单任务基线，并且学习的路由策略提供了对模态相关性和任务结构的可解释见解。这解决了个性化医疗保健的关键挑战，使每个主题的自适应信息处理，占数据的异质性和任务的相关性。应用于心理治疗，这一框架可以改善心理健康的结果，提高治疗分配的准确性，并通过个性化的干预策略，提高临床成本效益。
摘要：We propose a unified framework for adaptive routing in multitask, multimodal prediction settings where data heterogeneity and task interactions vary across samples. Motivated by applications in psychotherapy where structured assessments and unstructured clinician notes coexist with partially missing data and correlated outcomes, we introduce a routing-based architecture that dynamically selects modality processing pathways and task-sharing strategies on a per-sample basis. Our model defines multiple modality paths, including raw and fused representations of text and numeric features and learns to route each input through the most informative expert combination. Task-specific predictions are produced by shared or independent heads depending on the routing decision, and the entire system is trained end-to-end. We evaluate the model on both synthetic data and real-world psychotherapy notes predicting depression and anxiety outcomes. Our experiments show that our method consistently outperforms fixed multitask or single-task baselines, and that the learned routing policy provides interpretable insights into modality relevance and task structure. This addresses critical challenges in personalized healthcare by enabling per-subject adaptive information processing that accounts for data heterogeneity and task correlations. Applied to psychotherapy, this framework could improve mental health outcomes, enhance treatment assignment precision, and increase clinical cost-effectiveness through personalized intervention strategies.
强化学习(5篇)
【1】WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning
标题：WebSailor-V2：通过合成数据和可扩展的强化学习弥合与专有代理的鸿沟
链接：https://arxiv.org/abs/2509.13305
备注：his https URL
摘要：超越人类的认知局限性是LLM培训的关键前沿。像DeepResearch这样的专有代理系统已经在极其复杂的信息搜索基准上展示了超人的能力，例如BrowseComp，这是以前无法实现的壮举。我们认为，他们的成功取决于开源模型中缺乏的复杂推理模式：在浏览大量信息时系统地减少极端不确定性的能力。基于这一见解，我们引入了WebSailor，这是一种旨在灌输这一关键能力的完整的培训后方法。我们的方法包括通过结构化采样和信息混淆，RFT冷启动以及高效的代理RL训练算法，重复采样策略优化（DUPO）来生成新颖的，高不确定性的任务。通过这种集成管道，WebSailor在复杂的信息搜索任务中的表现明显优于所有开源代理，与专有代理的性能相匹配，并缩小了能力差距。
摘要：Transcending human cognitive limitations represents a critical frontier in LLM training. Proprietary agentic systems like DeepResearch have demonstrated superhuman capabilities on extremely complex information-seeking benchmarks such as BrowseComp, a feat previously unattainable. We posit that their success hinges on a sophisticated reasoning pattern absent in open-source models: the ability to systematically reduce extreme uncertainty when navigating vast information landscapes. Based on this insight, we introduce WebSailor, a complete post-training methodology designed to instill this crucial capability. Our approach involves generating novel, high-uncertainty tasks through structured sampling and information obfuscation, RFT cold start, and an efficient agentic RL training algorithm, Duplicating Sampling Policy Optimization (DUPO). With this integrated pipeline, WebSailor significantly outperforms all open-source agents in complex information-seeking tasks, matching proprietary agents' performance and closing the capability gap.
【2】Tool-R1: Sample-Efficient Reinforcement Learning for Agentic Tool Use
标题：Tools-R1：用于抽象工具使用的样本高效强化学习
链接：https://arxiv.org/abs/2509.12867
摘要：大型语言模型（LLM）在语言理解和推理方面表现出了强大的能力，但在处理需要最新知识、精确操作或专用工具使用的现实任务时，它们仍然有限。为了解决这个问题，我们提出了Tool-R1，这是一个强化学习框架，它使LLM能够通过生成可执行的Python代码来执行一般、组合和多步骤的工具使用。Tool-R1支持用户定义的工具和标准库的集成，跨步骤共享变量以构建一致的工作流程。基于结果的奖励功能，结合基于LLM的答案判断和代码执行成功，指导策略优化。为了提高训练效率，我们维护了一个动态的样本队列来缓存和重用高质量的轨迹，减少了昂贵的在线采样的开销。GAIA基准测试的实验表明，Tool-R1大大提高了准确性和鲁棒性，比强基线提高了约10%，并且在复杂的多步任务上有更大的改进。这些结果突出了Tool-R1在现实世界应用中实现可靠和高效的工具增强推理的潜力。我们的代码将在https://github.com/YBYBZhang/Tool-R1上提供。
摘要：Large language models (LLMs) have demonstrated strong capabilities in language understanding and reasoning, yet they remain limited when tackling real-world tasks that require up-to-date knowledge, precise operations, or specialized tool use. To address this, we propose Tool-R1, a reinforcement learning framework that enables LLMs to perform general, compositional, and multi-step tool use by generating executable Python code. Tool-R1 supports integration of user-defined tools and standard libraries, with variable sharing across steps to construct coherent workflows. An outcome-based reward function, combining LLM-based answer judgment and code execution success, guides policy optimization. To improve training efficiency, we maintain a dynamic sample queue to cache and reuse high-quality trajectories, reducing the overhead of costly online sampling. Experiments on the GAIA benchmark show that Tool-R1 substantially improves both accuracy and robustness, achieving about 10\% gain over strong baselines, with larger improvements on complex multi-step tasks. These results highlight the potential of Tool-R1 for enabling reliable and efficient tool-augmented reasoning in real-world applications. Our code will be available at https://github.com/YBYBZhang/Tool-R1.
【3】Safe Reinforcement Learning using Action Projection: Safeguard the Policy or the Environment?
标题：使用行动投射的安全强化学习：保护政策还是环境？
链接：https://arxiv.org/abs/2509.12833
摘要：基于投影的安全过滤器通过将不安全行为映射到最接近的安全替代方案来修改不安全行为，被广泛用于强化学习（RL）中的安全约束。通常考虑两种集成策略：安全环境RL（SE-RL），其中保护措施被视为环境的一部分，以及安全策略RL（SP-RL），其中它通过可区分的优化层嵌入策略中。尽管它们在安全关键环境中具有实际意义，但对它们的差异缺乏正式的了解。在这项工作中，我们提出了一个理论比较SE-RL和SP-RL。我们确定了一个关键的区别，每种方法是如何受到行动别名，多个不安全的行动被投射到同一个安全的行动，导致信息丢失的政策梯度的现象。在SE-RL中，这种效应被批评者隐含地近似，而在SP-RL中，它直接表现为通过保障措施的反向传播过程中的秩亏雅可比矩阵。我们的贡献有三个方面：（i）在行动者-批评者算法的背景下，SE-RL和SP-RL的统一形式化，（ii）对它们各自的政策梯度估计的理论分析，突出行动混淆的作用，以及（iii）缓解策略的比较研究，包括与既定SE-RL实践相一致的SP-RL的基于惩罚的改进。实证结果支持我们的理论预测，表明动作混叠是更有害的SP-RL比SE-RL。然而，通过适当的改进策略，SP-RL可以在一系列环境中匹配或优于改进的SE-RL。这些发现为基于任务特征选择和改进基于投影的安全RL方法提供了可操作的见解。
摘要：Projection-based safety filters, which modify unsafe actions by mapping them to the closest safe alternative, are widely used to enforce safety constraints in reinforcement learning (RL). Two integration strategies are commonly considered: Safe environment RL (SE-RL), where the safeguard is treated as part of the environment, and safe policy RL (SP-RL), where it is embedded within the policy through differentiable optimization layers. Despite their practical relevance in safety-critical settings, a formal understanding of their differences is lacking. In this work, we present a theoretical comparison of SE-RL and SP-RL. We identify a key distinction in how each approach is affected by action aliasing, a phenomenon in which multiple unsafe actions are projected to the same safe action, causing information loss in the policy gradients. In SE-RL, this effect is implicitly approximated by the critic, while in SP-RL, it manifests directly as rank-deficient Jacobians during backpropagation through the safeguard. Our contributions are threefold: (i) a unified formalization of SE-RL and SP-RL in the context of actor-critic algorithms, (ii) a theoretical analysis of their respective policy gradient estimates, highlighting the role of action aliasing, and (iii) a comparative study of mitigation strategies, including a novel penalty-based improvement for SP-RL that aligns with established SE-RL practices. Empirical results support our theoretical predictions, showing that action aliasing is more detrimental for SP-RL than for SE-RL. However, with appropriate improvement strategies, SP-RL can match or outperform improved SE-RL across a range of environments. These findings provide actionable insights for choosing and refining projection-based safe RL methods based on task characteristics.
【4】Pre-trained Visual Representations Generalize Where it Matters in Model-Based Reinforcement Learning
标题：预训练的视觉表示概括了基于模型的强化学习中的重要性
链接：https://arxiv.org/abs/2509.12531
摘要：在视觉策略学习中，机器人智能体的控制策略直接来自视觉输入。典型的方法是从头开始联合训练策略和视觉编码器，这种方法对新颖的视觉场景变化的推广效果很差。使用预训练的视觉模型（PVM）来通知策略网络，提高了无模型强化学习（MFRL）的鲁棒性。基于模型的强化学习（MBRL）的最新发展表明MBRL比MFRL更有效。然而，与直觉相反，现有的工作已经发现PVM在MBRL中是无效的。在这里，我们研究PVM在MBRL中的有效性，特别是在视觉域变化下的泛化方面。我们表明，在严重变化的情况下，PVM的表现要比从头开始训练的基线模型好得多。我们进一步研究了PVM微调的不同水平的影响。我们的研究结果表明，部分微调可以保持最高的平均任务性能下最极端的分布变化。我们的研究结果表明，PVM在促进视觉策略学习的鲁棒性方面非常成功，为它们在基于模型的机器人学习应用中的广泛采用提供了令人信服的证据。
摘要 ：In visuomotor policy learning, the control policy for the robotic agent is derived directly from visual inputs. The typical approach, where a policy and vision encoder are trained jointly from scratch, generalizes poorly to novel visual scene changes. Using pre-trained vision models (PVMs) to inform a policy network improves robustness in model-free reinforcement learning (MFRL). Recent developments in Model-based reinforcement learning (MBRL) suggest that MBRL is more sample-efficient than MFRL. However, counterintuitively, existing work has found PVMs to be ineffective in MBRL. Here, we investigate PVM's effectiveness in MBRL, specifically on generalization under visual domain shifts. We show that, in scenarios with severe shifts, PVMs perform much better than a baseline model trained from scratch. We further investigate the effects of varying levels of fine-tuning of PVMs. Our results show that partial fine-tuning can maintain the highest average task performance under the most extreme distribution shifts. Our results demonstrate that PVMs are highly successful in promoting robustness in visual policy learning, providing compelling evidence for their wider adoption in model-based robotic learning applications.
【5】Research on Short-Video Platform User Decision-Making via Multimodal Temporal Modeling and Reinforcement Learning
标题：基于多模式时态建模和强化学习的短视频平台用户决策研究
链接：https://arxiv.org/abs/2509.12269
备注：26 pages
摘要：本文提出了MT-DQN模型，该模型集成了Transformer、时间图神经网络（TGNN）和深度Q网络（DQN），以解决短视频环境中预测用户行为和优化推荐策略的挑战。实验表明，MT-DQN始终优于传统的级联模型，如Concat-Modal，平均F1分数提高了10.97%，平均NDCG@5提高了8.3%。与经典的强化学习模型Vanilla-DQN相比，MT-DQN将MSE降低了34.8%，MAE降低了26.5%。尽管如此，我们也认识到在现实场景中部署MT-DQN的挑战，例如在线推理期间的计算成本和延迟敏感性，这将通过未来的架构优化来解决。
摘要：This paper proposes the MT-DQN model, which integrates a Transformer, Temporal Graph Neural Network (TGNN), and Deep Q-Network (DQN) to address the challenges of predicting user behavior and optimizing recommendation strategies in short-video environments. Experiments demonstrated that MT-DQN consistently outperforms traditional concatenated models, such as Concat-Modal, achieving an average F1-score improvement of 10.97% and an average NDCG@5 improvement of 8.3%. Compared to the classic reinforcement learning model Vanilla-DQN, MT-DQN reduces MSE by 34.8% and MAE by 26.5%. Nonetheless, we also recognize challenges in deploying MT-DQN in real-world scenarios, such as its computational cost and latency sensitivity during online inference, which will be addressed through future architectural optimization.
元学习(1篇)
【1】Meta-model Neural Process for Probabilistic Power Flow under Varying N-1 System Topologies
标题：变化N-1系统布局下概率潮流的元模型神经过程
链接：https://arxiv.org/abs/2509.12281
备注：An improved version for the conference paper at PESGM 2025
摘要：概率潮流（PPF）问题是量化由于不确定注入的节点电压分布的基本。传统的PPF问题考虑的是一个固定的拓扑结构，并且这种PPF问题的解决方案与该拓扑结构相关联。拓扑结构的变化可能会改变潮流模式，因此需要再次解决PPF问题。以前的PPF模型及其解决方案不再适用于新的拓扑结构。这种做法带来了不便和计算负担，因为由于高可再生能源和电动汽车的大份额，预计会出现更多的意外情况。本文提出了一种新的拓扑自适应方法，基于元模型神经过程（MMNP），寻找不同的N-1拓扑下的PPF问题的解决方案，特别是与一线故障。通过利用基于上下文集的拓扑表示和条件分布函数学习技术，建议MMNP增强了PPF模型对拓扑变化的鲁棒性，减轻了在新配置上重新训练PPF模型的需要。对IEEE 9节点系统和IEEE 118节点系统的仿真验证了模型的有效性。在9总线和118总线中观察到的最大%L1相对误差范数分别为1.11%和0.77%。这种自适应方法填补了电网波动性不断增加的时代中PPF方法的关键空白。
摘要：The probabilistic power flow (PPF) problem is essential to quantifying the distribution of the nodal voltages due to uncertain injections. The conventional PPF problem considers a fixed topology, and the solutions to such a PPF problem are associated with this topology. A change in the topology might alter the power flow patterns and thus require the PPF problem to be solved again. The previous PPF model and its solutions are no longer valid for the new topology. This practice incurs both inconvenience and computation burdens as more contingencies are foreseen due to high renewables and a large share of electric vehicles. This paper presents a novel topology-adaptive approach, based on the meta-model Neural Process (MMNP), for finding the solutions to PPF problems under varying N-1 topologies, particularly with one-line failures. By leveraging context set-based topology representation and conditional distribution over function learning techniques, the proposed MMNP enhances the robustness of PPF models to topology variations, mitigating the need for retraining PPF models on a new configuration. Simulations on an IEEE 9-bus system and IEEE 118-bus system validate the model's performance. The maximum %L1-relative error norm was observed as 1.11% and 0.77% in 9-bus and 118-bus, respectively. This adaptive approach fills a critical gap in PPF methodology in an era of increasing grid volatility.
符号|符号学习(1篇)
【1】A Traditional Approach to Symbolic Piano Continuation
标题：象征性钢琴延续的传统方法
链接：https://arxiv.org/abs/2509.12267
备注：3 pages, extended abstract, MIREX session at ISMIR 2025 LBD
摘要：我们为MIREX 2025象征音乐世代挑战赛提供了一种传统的象征钢琴音乐延续方法。虽然计算音乐生成最近专注于开发大型基础模型与复杂的架构修改，我们认为，更简单的方法仍然更有效的约束，单乐器的任务。因此，我们回到了一个简单的，未增强的下一个令牌预测目标，目标是通过使用更好的数据和更好的基本面来超越大型基础模型。我们在https://github.com/christianazinn/mirex2025上发布模型重量和代码。
摘要：We present a traditional approach to symbolic piano music continuation for the MIREX 2025 Symbolic Music Generation challenge. While computational music generation has recently focused on developing large foundation models with sophisticated architectural modifications, we argue that simpler approaches remain more effective for constrained, single-instrument tasks. We thus return to a simple, unaugmented next-token-prediction objective on tokenized raw MIDI, aiming to outperform large foundation models by using better data and better fundamentals. We release model weights and code at https://github.com/christianazinn/mirex2025.
分层学习(1篇)
【1】HAM: Hierarchical Adapter Merging for Scalable Continual Learning
标题：HAM：分层适配器合并以实现可扩展的持续学习
链接：https://arxiv.org/abs/2509.13211
摘要 ：持续学习是人类认知的基本能力，但它对当前的深度学习模型提出了重大挑战。主要的问题是，新知识可能会干扰以前学习的信息，导致模型忘记以前的知识，而倾向于新知识，这种现象被称为灾难性遗忘。虽然大型预训练模型可以通过利用其现有知识和过度参数化来部分减轻遗忘，但当面对新的数据分布时，它们往往会遇到困难。参数高效微调（PEFT）方法，如LoRA，可以有效地适应新知识。然而，它们在扩展到动态学习场景和长序列任务方面仍然面临挑战，因为每个任务维护一个适配器会引入复杂性并增加干扰的可能性。在本文中，我们介绍了分层适配器合并（HAM），这是一种新颖的框架，可以在训练期间动态组合来自不同任务的适配器。这种方法使HAM能够有效地扩展，使其能够以更高的效率管理比竞争基线更多的任务。为了实现这一点，HAM维护了一组固定的组，这些组按层次结构合并新的适配器。对于每个任务，HAM训练低等级适配器以及重要性标量，然后基于适配器相似性动态分组任务。在每个组中，适配器被修剪，缩放和合并，促进相关任务之间的迁移学习。在三个视觉基准上进行的大量实验表明，HAM的性能明显优于最先进的方法，特别是随着任务数量的增加。
摘要：Continual learning is an essential capability of human cognition, yet it poses significant challenges for current deep learning models. The primary issue is that new knowledge can interfere with previously learned information, causing the model to forget earlier knowledge in favor of the new, a phenomenon known as catastrophic forgetting. Although large pre-trained models can partially mitigate forgetting by leveraging their existing knowledge and over-parameterization, they often struggle when confronted with novel data distributions. Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, enable efficient adaptation to new knowledge. However, they still face challenges in scaling to dynamic learning scenarios and long sequences of tasks, as maintaining one adapter per task introduces complexity and increases the potential for interference. In this paper, we introduce Hierarchical Adapters Merging (HAM), a novel framework that dynamically combines adapters from different tasks during training. This approach enables HAM to scale effectively, allowing it to manage more tasks than competing baselines with improved efficiency. To achieve this, HAM maintains a fixed set of groups that hierarchically consolidate new adapters. For each task, HAM trains a low-rank adapter along with an importance scalar, then dynamically groups tasks based on adapter similarity. Within each group, adapters are pruned, scaled and merge, facilitating transfer learning between related tasks. Extensive experiments on three vision benchmarks show that HAM significantly outperforms state-of-the-art methods, particularly as the number of tasks increases.
医学相关(7篇)
【1】NORA: A Nephrology-Oriented Representation Learning Approach Towards Chronic Kidney Disease Classification
标题：NORA：一种面向肾病的代表学习方法，用于慢性肾病分类
链接：https://arxiv.org/abs/2509.12704
备注：7 pages, 5 figures, accepted to the International Conference on Machine Learning and Applications (ICMLA) 2025
摘要：慢性肾脏病（CKD）影响着全球数百万人，但其早期检测仍然具有挑战性，特别是在门诊环境中，通常无法获得基于实验室的肾脏生物标志物。在这项工作中，我们研究了常规收集的非肾脏临床变量对CKD分类的预测潜力，包括社会人口因素，共病情况和尿分析结果。我们介绍了面向肾脏的表示学习（NORA）方法，该方法将监督对比学习与非线性随机森林分类器相结合。NORA首先从表格EHR数据中导出有区别的患者表示，然后将其用于下游CKD分类。我们在Riverside Nephrology Physicians基于临床的EHR数据集上评估了NORA。我们的研究结果表明，NORA提高了类可分性和整体分类性能，特别是提高了早期CKD的F1评分。此外，我们评估了NORA在UCI CKD数据集上的普遍性，证明了其在不同患者队列中对CKD风险分层的有效性。
摘要：Chronic Kidney Disease (CKD) affects millions of people worldwide, yet its early detection remains challenging, especially in outpatient settings where laboratory-based renal biomarkers are often unavailable. In this work, we investigate the predictive potential of routinely collected non-renal clinical variables for CKD classification, including sociodemographic factors, comorbid conditions, and urinalysis findings. We introduce the Nephrology-Oriented Representation leArning (NORA) approach, which combines supervised contrastive learning with a nonlinear Random Forest classifier. NORA first derives discriminative patient representations from tabular EHR data, which are then used for downstream CKD classification. We evaluated NORA on a clinic-based EHR dataset from Riverside Nephrology Physicians. Our results demonstrated that NORA improves class separability and overall classification performance, particularly enhancing the F1-score for early-stage CKD. Additionally, we assessed the generalizability of NORA on the UCI CKD dataset, demonstrating its effectiveness for CKD risk stratification across distinct patient cohorts.
【2】A Multimodal Foundation Model to Enhance Generalizability and Data Efficiency for Pan-cancer Prognosis Prediction
标题：提高泛癌症预后预测的概括性和数据效率的多峰基础模型
链接：https://arxiv.org/abs/2509.12600
备注：27 pages, 7 figures
摘要：多模态数据为全面了解肿瘤微环境提供了异构信息。然而，现有的人工智能模型往往难以利用多模态数据中的丰富信息，并提取出难以推广的表示。在这里，我们提出了MICE（通过协作专家进行多模式数据集成），这是一种多模式基础模型，可有效集成病理图像，临床报告和基因组学数据，以进行精确的泛癌症预后预测。与传统的多专家模块不同，MICE采用多个功能多样化的专家来全面捕获跨癌症和癌症特异性的见解。利用来自30种癌症类型的11，799名患者的数据，我们通过结合对比学习和监督学习来增强MICE的可推广性。MICE的表现优于单峰模型和最先进的基于多专家的多模式模型，表明内部队列的C指数大幅改善，分别为3.8%至11.2%和5.8%至8.8%。此外，它在各种临床场景中表现出显着的数据效率。凭借其增强的可推广性和数据效率，MICE为泛癌症预后预测建立了有效和可扩展的基础，具有个性化定制治疗和改善治疗结果的强大潜力。
摘要：Multimodal data provides heterogeneous information for a holistic understanding of the tumor microenvironment. However, existing AI models often struggle to harness the rich information within multimodal data and extract poorly generalizable representations. Here we present MICE (Multimodal data Integration via Collaborative Experts), a multimodal foundation model that effectively integrates pathology images, clinical reports, and genomics data for precise pan-cancer prognosis prediction. Instead of conventional multi-expert modules, MICE employs multiple functionally diverse experts to comprehensively capture both cross-cancer and cancer-specific insights. Leveraging data from 11,799 patients across 30 cancer types, we enhanced MICE's generalizability by coupling contrastive and supervised learning. MICE outperformed both unimodal and state-of-the-art multi-expert-based multimodal models, demonstrating substantial improvements in C-index ranging from 3.8% to 11.2% on internal cohorts and 5.8% to 8.8% on independent cohorts, respectively. Moreover, it exhibited remarkable data efficiency across diverse clinical scenarios. With its enhanced generalizability and data efficiency, MICE establishes an effective and scalable foundation for pan-cancer prognosis prediction, holding strong potential to personalize tailored therapies and improve treatment outcomes.
【3】Developing an aeroponic smart experimental greenhouse for controlling irrigation and plant disease detection using deep learning and IoT
标题：利用深度学习和物联网开发用于控制灌溉和植物病害检测的气培智能实验温室
链接：https://arxiv.org/abs/2509.12274
备注：Author-accepted version. Presented at ASABE Annual International Meeting (AIM) 2021 (virtual), Paper 2101252. Please cite the published meeting paper: doi:10.13031/aim.202101252. Minor wording and formatting updates in this preprint
摘要 ：控制环境条件和监测温室中的植物状况对于及时做出旨在促进作物生产的适当管理决策至关重要。这项研究的主要目标是在实验规模上开发和测试智能气培温室，通过物联网（IoT）和人工智能（AI）的整合，持续监测植物的状态和环境条件。开发了一个基于物联网的平台，以更有效地控制工厂的环境条件，并为用户提供见解，以做出明智的管理决策。此外，我们开发了一个基于AI的疾病检测框架，使用VGG-19、InceptionResNetV 2和InceptionV 3算法来分析在有意接种后定期捕获的图像。AI框架的性能与专家对疾病状态的评估进行了比较。初步结果表明，在温室环境中实施的物联网系统能够连续向用户在线发布温度、湿度、水流量和充电罐体积等数据，并调整控制参数，为植物提供最佳生长环境。此外，人工智能框架的结果表明，VGG-19算法能够以最高的准确率从健康叶片中识别干旱胁迫和锈病叶片，在其他算法中达到92%。
摘要：Controlling environmental conditions and monitoring plant status in greenhouses is critical to promptly making appropriate management decisions aimed at promoting crop production. The primary objective of this research study was to develop and test a smart aeroponic greenhouse on an experimental scale where the status of Geranium plant and environmental conditions are continuously monitored through the integration of the internet of things (IoT) and artificial intelligence (AI). An IoT-based platform was developed to control the environmental conditions of plants more efficiently and provide insights to users to make informed management decisions. In addition, we developed an AI-based disease detection framework using VGG-19, InceptionResNetV2, and InceptionV3 algorithms to analyze the images captured periodically after an intentional inoculation. The performance of the AI framework was compared with an expert's evaluation of disease status. Preliminary results showed that the IoT system implemented in the greenhouse environment is able to publish data such as temperature, humidity, water flow, and volume of charge tanks online continuously to users and adjust the controlled parameters to provide an optimal growth environment for the plants. Furthermore, the results of the AI framework demonstrate that the VGG-19 algorithm was able to identify drought stress and rust leaves from healthy leaves with the highest accuracy, 92% among the other algorithms.
【4】Interpretable Data Mining of Follicular Thyroid Cancer Ultrasound Features Using Enhanced Association Rules
标题：使用增强型关联规则对甲状腺毛囊癌超声特征进行可解释数据挖掘
链接：https://arxiv.org/abs/2509.12238
摘要：目的：甲状腺癌是一种常见的恶性肿瘤。甲状腺癌的早期症状有哪些？甲状腺癌的早期症状有哪些？滤泡性甲状腺癌缺乏独特的超声体征，比更普遍的乳头状甲状腺癌更难在术前诊断，而且与之相关的临床研究也不太成熟。我们的目的是分析滤泡性甲状腺癌的临床数据的基础上，一种新的数据挖掘工具，以确定一些临床指征，可能有助于术前诊断。研究方法：我们对北京大学第三医院普外科2010年至2023年收集的病例数据进行了回顾性分析。与传统的统计方法不同，我们改进了经典的数据挖掘方法关联规则挖掘，并借助可解释机器学习中SHAP方法的思想，提出了新的反映临床指征与癌症之间恶性关联的分析指标。结果如下：该数据集经过预处理，包含1673例病例（根据淋巴结而不是患者），其中1414例为良性淋巴结，259例为恶性淋巴结。我们的分析指出，除了一些常见的指标（例如，结节边缘不规则或分叶状、厚度不均匀晕、低回声），也有一些与恶性相关性强的指标，如结节中结节型、小梁型和低TSH评分。此外，我们的研究结果表明，桥本甲状腺炎的组合也可能有很强的恶性协会。结论：在术前诊断疑似甲状腺滤泡癌的结节时，应考虑多种临床指征，以获得更准确的诊断。本研究所发现的多种恶性相关性可作为相关领域临床医师的参考。
摘要：Purpose: Thyroid cancer has been a common cancer. Papillary thyroid cancer and follicular thyroid cancer are the two most common types of thyroid cancer. Follicular thyroid cancer lacks distinctive ultrasound signs and is more difficult to diagnose preoperatively than the more prevalent papillary thyroid cancer, and the clinical studies associated with it are less well established. We aimed to analyze the clinical data of follicular thyroid cancer based on a novel data mining tool to identify some clinical indications that may help in preoperative diagnosis. Methods: We performed a retrospective analysis based on case data collected by the Department of General Surgery of Peking University Third Hospital between 2010 and 2023. Unlike traditional statistical methods, we improved the association rule mining, a classical data mining method, and proposed new analytical metrics reflecting the malignant association between clinical indications and cancer with the help of the idea of SHAP method in interpretable machine learning. Results: The dataset was preprocessed to contain 1673 cases (in terms of nodes rather than patients), of which 1414 were benign and 259 were malignant nodes. Our analysis pointed out that in addition to some common indicators (e.g., irregular or lobulated nodal margins, uneven thickness halo, hypoechogenicity), there were also some indicators with strong malignant associations, such as nodule-in-nodule pattern, trabecular pattern, and low TSH scores. In addition, our results suggest that the combination of Hashimoto's thyroiditis may also have a strong malignant association. Conclusion: In the preoperative diagnosis of nodules suspected of follicular thyroid cancer, multiple clinical indications should be considered for a more accurate diagnosis. The diverse malignant associations identified in our study may serve as a reference for clinicians in related fields.
【5】Flexible Multimodal Neuroimaging Fusion for Alzheimer's Disease Progression Prediction
标题：灵活的多模式神经影像融合用于阿尔茨海默病进展预测
链接：https://arxiv.org/abs/2509.12234
备注：Accepted at Applications of Medical AI 2025
摘要：阿尔茨海默病（AD）是一种进行性神经退行性疾病，患者之间的认知能力下降率差异很大。AD进展预测的目的是预测患者的认知能力下降，并从多种神经影像学方式中获益。然而，现有的多模态模型无法做出准确的预测时，许多模态在推理过程中丢失，这是经常在临床环境中的情况。为了增加多模态模型的灵活性下，高模态缺失，我们引入PerM-MoE，一种新的稀疏混合专家的方法，使用独立的路由器为每一个模态的传统，单一的路由器。使用T1加权MRI、FLAIR、淀粉样蛋白β PET和来自阿尔茨海默病神经成像倡议（ADNI）的tau PET神经成像数据，我们评估了PerM-MoE、最先进的Flex-MoE和单峰神经成像模型在不同模态缺失水平下预测临床痴呆评分总和（CDR-SB）评分的两年变化。PerM-MoE在模态缺失的大多数变化中优于最先进的技术水平，并证明了比Flex-MoE更有效的专家效用。
摘要：Alzheimer's disease (AD) is a progressive neurodegenerative disease with high inter-patient variance in rate of cognitive decline. AD progression prediction aims to forecast patient cognitive decline and benefits from incorporating multiple neuroimaging modalities. However, existing multimodal models fail to make accurate predictions when many modalities are missing during inference, as is often the case in clinical settings. To increase multimodal model flexibility under high modality missingness, we introduce PerM-MoE, a novel sparse mixture-of-experts method that uses independent routers for each modality in place of the conventional, single router. Using T1-weighted MRI, FLAIR, amyloid beta PET, and tau PET neuroimaging data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), we evaluate PerM-MoE, state-of-the-art Flex-MoE, and unimodal neuroimaging models on predicting two-year change in Clinical Dementia Rating-Sum of Boxes (CDR-SB) scores under varying levels of modality missingness. PerM-MoE outperforms the state of the art in most variations of modality missingness and demonstrates more effective utility of experts than Flex-MoE.
【6】Enhancing Radiographic Disease Detection with MetaCheX, a Context-Aware Multimodal Model
标题：使用MetaCheX（一种上下文感知多模式模型）增强放射学疾病检测
链接：https://arxiv.org/abs/2509.12287
备注：All authors contributed equally, 5 pages, 2 figures, 1 table
摘要：现有的胸部放射学深度学习模型往往忽视患者元数据，限制了诊断的准确性和公平性。为了弥合这一差距，我们引入了MetaCheX，这是一种新型的多模式框架，它将胸部X射线图像与结构化的患者元数据集成在一起，以复制临床决策。我们的方法结合了卷积神经网络（CNN）的骨干与元数据处理的多层感知器通过一个共享的分类。在CheXpert Plus数据集上进行评估，MetaCheX在多个CNN架构中的表现始终优于仅射线照相基线模型。通过整合元数据，整体诊断准确性显著提高，通过AUROC的增加来衡量。这项研究的结果表明，元数据减少了算法偏差，并提高了模型在不同患者人群中的通用性。MetaCheX将临床人工智能推进到强大的、上下文感知的放射学疾病检测。
摘要 ：Existing deep learning models for chest radiology often neglect patient metadata, limiting diagnostic accuracy and fairness. To bridge this gap, we introduce MetaCheX, a novel multimodal framework that integrates chest X-ray images with structured patient metadata to replicate clinical decision-making. Our approach combines a convolutional neural network (CNN) backbone with metadata processed by a multilayer perceptron through a shared classifier. Evaluated on the CheXpert Plus dataset, MetaCheX consistently outperformed radiograph-only baseline models across multiple CNN architectures. By integrating metadata, the overall diagnostic accuracy was significantly improved, measured by an increase in AUROC. The results of this study demonstrate that metadata reduces algorithmic bias and enhances model generalizability across diverse patient populations. MetaCheX advances clinical artificial intelligence toward robust, context-aware radiographic disease detection.
【7】CNN-BiLSTM for sustainable and non-invasive COVID-19 detection via salivary ATR-FTIR spectroscopy
标题：CNN-BiLSTM通过唾液ATR-FTIR光谱进行可持续且非侵入性的COVID-19检测
链接：https://arxiv.org/abs/2509.12241
摘要：COVID-19大流行对医疗保健系统造成前所未有的压力，并仍然是全球健康问题，特别是随着新变种的出现。尽管实时聚合酶链反应（RT-PCR）被认为是COVID-19检测的金标准，但它昂贵、耗时、劳动密集，并且对RNA提取问题敏感。在这种情况下，生物流体的ATR-FTIR光谱分析为COVID-19检测提供了一种无试剂、具有成本效益的替代方案。我们提出了一种新的架构，将卷积神经网络（CNN）与双向长短期记忆（BiLSTM）网络相结合，称为CNN-BiLSTM，以处理ATR-FTIR光谱生成的光谱并从光谱样本中诊断COVID-19。我们将这种架构的性能与独立的CNN和其他最先进的机器学习技术进行了比较。实验结果表明，我们的CNN-BiLSTM模型优于所有其他模型，在具有挑战性的现实世界COVID-19数据集上实现了平均准确率和0.80的F1得分。在CNN架构中加入BiLSTM层，显著增强了模型性能，使CNN-BiLSTM成为使用非侵入性唾液样本的ATR-FTIR光谱检测COVID-19的更准确可靠的选择。
摘要：The COVID-19 pandemic has placed unprecedented strain on healthcare systems and remains a global health concern, especially with the emergence of new variants. Although real-time polymerase chain reaction (RT-PCR) is considered the gold standard for COVID-19 detection, it is expensive, time-consuming, labor-intensive, and sensitive to issues with RNA extraction. In this context, ATR-FTIR spectroscopy analysis of biofluids offers a reagent-free, cost-effective alternative for COVID-19 detection. We propose a novel architecture that combines Convolutional Neural Networks (CNN) with Bidirectional Long Short-Term Memory (BiLSTM) networks, referred to as CNN-BiLSTM, to process spectra generated by ATR-FTIR spectroscopy and diagnose COVID-19 from spectral samples. We compare the performance of this architecture against a standalone CNN and other state-of-the-art machine learning techniques. Experimental results demonstrate that our CNN-BiLSTM model outperforms all other models, achieving an average accuracy and F1-score of 0.80 on a challenging real-world COVID-19 dataset. The addition of the BiLSTM layer to the CNN architecture significantly enhances model performance, making CNN-BiLSTM a more accurate and reliable choice for detecting COVID-19 using ATR-FTIR spectra of non-invasive saliva samples.
蒸馏|知识提取(3篇)
【1】iCD: A Implicit Clustering Distillation Mathod for Structural Information Mining
标题：iCD：结构信息挖掘的隐式集群蒸馏方法
链接：https://arxiv.org/abs/2509.12553
摘要：Logit知识蒸馏近年来获得了大量的研究兴趣，由于其简单性和缺乏中间特征对齐的要求，但是，它在其决策过程中受到有限的解释性。为了解决这个问题，我们提出了隐式聚类蒸馏（iCD）：一种简单有效的方法，可以从logits中挖掘和传输可解释的结构知识，而不需要地面实况标签或特征空间对齐。iCD利用Gram矩阵在解耦的局部logit表示上，使学生模型能够学习潜在的语义结构模式。在基准数据集上进行的大量实验证明了iCD在不同师生架构中的有效性，在细粒度分类任务中表现尤其出色-比基线实现了+5.08%的峰值改进。该代码可从以下网址获得：https://github.com/maomaochongaa/iCD。
摘要：Logit Knowledge Distillation has gained substantial research interest in recent years due to its simplicity and lack of requirement for intermediate feature alignment; however, it suffers from limited interpretability in its decision-making process. To address this, we propose implicit Clustering Distillation (iCD): a simple and effective method that mines and transfers interpretable structural knowledge from logits, without requiring ground-truth labels or feature-space alignment. iCD leverages Gram matrices over decoupled local logit representations to enable student models to learn latent semantic structural patterns. Extensive experiments on benchmark datasets demonstrate the effectiveness of iCD across diverse teacher-student architectures, with particularly strong performance in fine-grained classification tasks -- achieving a peak improvement of +5.08% over the baseline. The code is available at: https://github.com/maomaochongaa/iCD.
【2】LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations
标题：LEAF：具有教师一致表示的文本嵌入模型的知识提炼
链接：https://arxiv.org/abs/2509.12539
备注：17 pages, 12 figures
摘要：我们提出了LEAF（“轻量级嵌入对齐框架”），文本嵌入模型的知识蒸馏框架。一个关键的区别特征是，我们提取的叶模型与它们的老师对齐。在信息检索的上下文中，这允许灵活的非对称架构，其中文档使用较大的教师模型编码，而查询可以使用较小的叶模型。我们还表明，叶模型自动继承MRL和输出量化的鲁棒性，只要这些属性存在于教师模型中，没有明确的训练。为了展示我们框架的能力，我们发布了leaf-ir，这是一个使用LEAF训练的23 M参数面向信息检索的文本嵌入模型，它在BEIR上设置了一个新的最先进的（SOTA），在该基准测试及其规模的模型的公共排行榜上排名第一。当运行在非对称模式下时，其检索性能进一步提高。然而，我们的计划是不限于信息检索设置，我们证明了其更广泛的适用性，通过合成多任务的叶MT模型。这还建立了一个新的SOTA，其规模在公共MTEB v2（英语）排行榜上排名第一。LEAF适用于黑盒模型，与其他嵌入模型训练框架相比，它不需要判断也不需要硬否定，并且可以使用小批量进行训练。因此，我们框架的数据集和培训基础设施要求并不高。我们在Apache 2.0许可证下公开提供我们的模型。
摘要：We present LEAF ("Lightweight Embedding Alignment Framework"), a knowledge distillation framework for text embedding models. A key distinguishing feature is that our distilled leaf models are aligned to their teacher. In the context of information retrieval, this allows for flexible asymmetric architectures where documents are encoded with the larger teacher model, while queries can be served with the smaller leaf models. We also show that leaf models automatically inherit MRL and robustness to output quantization whenever these properties are present in the teacher model, without explicitly training for them. To demonstrate the capability of our framework we publish leaf-ir, a 23M parameters information retrieval oriented text embedding model trained using LEAF, which sets a new state-of-the-art (SOTA) on BEIR, ranking #1 on the public leaderboard for this benchmark and for models of its size. When run in asymmetric mode, its retrieval performance is further increased. Our scheme is however not restricted to the information retrieval setting, and we demonstrate its wider applicability by synthesizing the multi-task leaf-mt model. This also sets a new SOTA, ranking #1 on the public MTEB v2 (English) leaderboard for its size. LEAF is applicable to black-box models and in contrast to other embedding model training frameworks, it does not require judgments nor hard negatives, and training can be conducted using small batch sizes. Thus, dataset and training infrastructure requirements for our framework are modest. We make our models publicly available under a permissive Apache 2.0 license.
【3】GhostNetV3-Small: A Tailored Architecture and Comparative Study of Distillation Strategies for Tiny Images
标题
：GhostNetV 3-Small：定制架构和微小图像蒸馏策略的比较研究
链接：https://arxiv.org/abs/2509.12380
摘要：深度神经网络在一系列任务中取得了显著的成功，但其计算需求往往使其不适合部署在资源受限的边缘设备上。本文探讨了压缩和调整模型的策略，以便在这种环境中进行有效的推理。我们专注于GhostNetV 3，一种最先进的移动应用程序架构，并提出了GhostNetV 3-Small，一种经过修改的变体，旨在更好地处理低分辨率输入，例如CIFAR-10数据集中的输入。除了架构适应，我们提供了一个比较评估的知识蒸馏技术，包括传统的知识蒸馏，教师助理，教师合奏。实验结果表明，GhostNetV 3-Small在CIFAR-10上的性能明显优于原始GhostNetV 3，准确率达到93.94%。与预期相反，与基线培训相比，所有检查的蒸馏策略都导致准确性降低。这些发现表明，在小规模图像分类任务中，架构自适应可能比蒸馏更有影响力，这突出表明需要进一步研究低分辨率域的有效模型设计和先进蒸馏技术。
摘要：Deep neural networks have achieved remarkable success across a range of tasks, however their computational demands often make them unsuitable for deployment on resource-constrained edge devices. This paper explores strategies for compressing and adapting models to enable efficient inference in such environments. We focus on GhostNetV3, a state-of-the-art architecture for mobile applications, and propose GhostNetV3-Small, a modified variant designed to perform better on low-resolution inputs such as those in the CIFAR-10 dataset. In addition to architectural adaptation, we provide a comparative evaluation of knowledge distillation techniques, including traditional knowledge distillation, teacher assistants, and teacher ensembles. Experimental results show that GhostNetV3-Small significantly outperforms the original GhostNetV3 on CIFAR-10, achieving an accuracy of 93.94%. Contrary to expectations, all examined distillation strategies led to reduced accuracy compared to baseline training. These findings indicate that architectural adaptation can be more impactful than distillation in small-scale image classification tasks, highlighting the need for further research on effective model design and advanced distillation techniques for low-resolution domains.
超分辨率|去噪|去模糊|去雾(1篇)
【1】ReTrack: Data Unlearning in Diffusion Models through Redirecting the Denoising Trajectory
标题：ReTrack：通过重定向去噪轨迹来消除扩散模型中的数据学习
链接：https://arxiv.org/abs/2509.13007
摘要：扩散模型擅长生成高质量、多样化的图像，但存在训练数据记忆问题，引发了严重的隐私和安全问题。数据非学习已经出现，通过消除特定数据的影响来缓解这个问题，而无需从头开始重新训练。我们提出了ReTrack，一个快速有效的扩散模型数据学习方法。ReTrack采用重要性抽样来构建更有效的微调损失，我们近似只保留主导条款。这产生了一个可解释的目标，将去噪轨迹重定向到$k$-最近邻，从而在保持生成质量的同时实现有效的非学习。在MNIST T-Shirt、CelebA-HQ、CIFAR-10和Stable Diffusion上的实验表明，ReTrack实现了最先进的性能，在遗忘强度和生成质量保持之间取得了最佳平衡。
摘要：Diffusion models excel at generating high-quality, diverse images but suffer from training data memorization, raising critical privacy and safety concerns. Data unlearning has emerged to mitigate this issue by removing the influence of specific data without retraining from scratch. We propose ReTrack, a fast and effective data unlearning method for diffusion models. ReTrack employs importance sampling to construct a more efficient fine-tuning loss, which we approximate by retaining only dominant terms. This yields an interpretable objective that redirects denoising trajectories toward the $k$-nearest neighbors, enabling efficient unlearning while preserving generative quality. Experiments on MNIST T-Shirt, CelebA-HQ, CIFAR-10, and Stable Diffusion show that ReTrack achieves state-of-the-art performance, striking the best trade-off between unlearning strength and generation quality preservation.
联邦学习|隐私保护|加密(4篇)
【1】Energy-Efficient Quantized Federated Learning for Resource-constrained IoT devices
标题：针对资源受限的物联网设备的节能量化联邦学习
链接：https://arxiv.org/abs/2509.12814
备注：6 pages, accepted at IEEE PIMRC 2025
摘要：联合学习（FL）已经成为一种很有前途的范例，可以实现协作机器学习，同时保护数据隐私，使其特别适合物联网（IoT）环境。然而，由于能量有限、通信信道不可靠以及假设无限块长度传输的不切实际性，资源受限的物联网设备面临着重大挑战。本文提出了一种物联网网络的联邦学习框架，该框架集成了有限块长度传输，模型量化和错误感知聚合机制，以提高能源效率和通信可靠性。该框架还优化了上行链路传输功率，以平衡节能和模型性能。仿真结果表明，与标准FL模型相比，所提出的方法可显著降低高达75%的能耗，同时保持强大的模型准确性，使其成为现实世界中资源受限的物联网场景中FL的可行解决方案。这项工作为实际物联网部署中高效可靠的FL实现铺平了道路。索引术语：联邦学习，物联网，有限块长度，量化，能源效率。
摘要：Federated Learning (FL) has emerged as a promising paradigm for enabling collaborative machine learning while preserving data privacy, making it particularly suitable for Internet of Things (IoT) environments. However, resource-constrained IoT devices face significant challenges due to limited energy,unreliable communication channels, and the impracticality of assuming infinite blocklength transmission. This paper proposes a federated learning framework for IoT networks that integrates finite blocklength transmission, model quantization, and an error-aware aggregation mechanism to enhance energy efficiency and communication reliability. The framework also optimizes uplink transmission power to balance energy savings and model performance. Simulation results demonstrate that the proposed approach significantly reduces energy consumption by up to 75\% compared to a standard FL model, while maintaining robust model accuracy, making it a viable solution for FL in real-world IoT scenarios with constrained resources. This work paves the way for efficient and reliable FL implementations in practical IoT deployments. Index Terms: Federated learning, IoT, finite blocklength, quantization, energy efficiency.
【2】High-Energy Concentration for Federated Learning in Frequency Domain
标题：频域联邦学习的高能集中
链接：https://arxiv.org/abs/2509.12630
摘要：联邦学习（FL）在没有数据共享的情况下为协作优化提供了巨大的潜力。由于合成数据被发送到服务器，利用流行的数据集蒸馏概念，该FL框架在减轻数据异构性的同时保护了真实数据的隐私。然而，这种方法仍然受到整个空间域设计中的冗余信息和噪声的挑战，这不可避免地增加了通信负担。在本文中，我们提出了一种新的频域感知FL方法与高能量浓度（FedFD）来解决这个问题。我们的FedFD的灵感来自于发现离散余弦变换主要将能量分布到特定区域，称为高能量集中。FedFD背后的原理是，低能量的高频分量通常包含冗余信息和噪声，因此过滤它们有助于降低通信成本并优化性能。我们的FedFD在数学上制定使用二进制掩码保存低频分量，通过频域分布对齐促进最佳解决方案。特别是，实际数据驱动的综合分类施加到损失，以提高低频分量的质量。在五个图像和语音数据集上，FedFD实现了比最先进方法更优越的性能，同时降低了通信成本。例如，在Dirichlet系数$\alpha = 0.01$的CIFAR-10数据集上，FedFD实现了通信成本的最小降低37.78\%，同时获得了10.88\%的性能增益。
摘要 ：Federated Learning (FL) presents significant potential for collaborative optimization without data sharing. Since synthetic data is sent to the server, leveraging the popular concept of dataset distillation, this FL framework protects real data privacy while alleviating data heterogeneity. However, such methods are still challenged by the redundant information and noise in entire spatial-domain designs, which inevitably increases the communication burden. In this paper, we propose a novel Frequency-Domain aware FL method with high-energy concentration (FedFD) to address this problem. Our FedFD is inspired by the discovery that the discrete cosine transform predominantly distributes energy to specific regions, referred to as high-energy concentration. The principle behind FedFD is that low-energy like high-frequency components usually contain redundant information and noise, thus filtering them helps reduce communication costs and optimize performance. Our FedFD is mathematically formulated to preserve the low-frequency components using a binary mask, facilitating an optimal solution through frequency-domain distribution alignment. In particular, real data-driven synthetic classification is imposed into the loss to enhance the quality of the low-frequency components. On five image and speech datasets, FedFD achieves superior performance than state-of-the-art methods while reducing communication costs. For example, on the CIFAR-10 dataset with Dirichlet coefficient $\alpha = 0.01$, FedFD achieves a minimum reduction of 37.78\% in the communication cost, while attaining a 10.88\% performance gain.
【3】Enhancing Smart Farming Through Federated Learning: A Secure, Scalable, and Efficient Approach for AI-Driven Agriculture
标题：通过联邦学习增强智能农业：一种安全，可扩展和高效的人工智能驱动农业方法
链接：https://arxiv.org/abs/2509.12363
备注：15 pages, 5 Figures
摘要：随着先进技术的整合，农业部门正在经历转型，特别是在数据驱动的决策方面。这项工作提出了一个智能农业的联合学习框架，旨在开发一个可扩展的，高效的和安全的解决方案，用于根据明尼苏达州农场的环境和运营条件进行作物病害检测。通过在本地维护敏感的农场数据并实现协作模型更新，我们提出的框架旨在实现作物病害分类的高准确性，而不会影响数据隐私。我们概述了一种方法，涉及从明尼苏达州农场收集数据，应用本地深度学习算法，迁移学习和用于模型优化的中央聚合服务器，旨在提高疾病检测的准确性，在农业场景中实现良好的泛化，降低通信和培训时间的成本，以及在未来实施中对疾病的早期识别和干预。我们概述了一种方法和预期的结果，为后续研究的实证验证奠定了基础。这项工作的背景是，越来越多的农业数据驱动解释的需求必须与对农场隐私的担忧相权衡，这些农场不愿意分享他们的运营数据。这对于提供一种安全有效的疾病检测方法至关重要，这种方法最终可以彻底改变智能农业系统，并在数据保密的情况下解决当地农业问题。在这样做的过程中，本文弥合了先进的机器学习技术与明尼苏达州及其他地区农民的实际，隐私敏感需求之间的差距，利用了联邦学习的好处。
摘要：The agricultural sector is undergoing a transformation with the integration of advanced technologies, particularly in data-driven decision-making. This work proposes a federated learning framework for smart farming, aiming to develop a scalable, efficient, and secure solution for crop disease detection tailored to the environmental and operational conditions of Minnesota farms. By maintaining sensitive farm data locally and enabling collaborative model updates, our proposed framework seeks to achieve high accuracy in crop disease classification without compromising data privacy. We outline a methodology involving data collection from Minnesota farms, application of local deep learning algorithms, transfer learning, and a central aggregation server for model refinement, aiming to achieve improved accuracy in disease detection, good generalization across agricultural scenarios, lower costs in communication and training time, and earlier identification and intervention against diseases in future implementations. We outline a methodology and anticipated outcomes, setting the stage for empirical validation in subsequent studies. This work comes in a context where more and more demand for data-driven interpretations in agriculture has to be weighed with concerns about privacy from farms that are hesitant to share their operational data. This will be important to provide a secure and efficient disease detection method that can finally revolutionize smart farming systems and solve local agricultural problems with data confidentiality. In doing so, this paper bridges the gap between advanced machine learning techniques and the practical, privacy-sensitive needs of farmers in Minnesota and beyond, leveraging the benefits of federated learning.
【4】Accelerating Privacy-Preserving Federated Learning in Large-Scale LEO Satellite Systems
标题：加速大规模LEO卫星系统中保护隐私的联邦学习
链接：https://arxiv.org/abs/2509.12222
备注：Submitted to IEEE conference for publication
摘要：大型低地球轨道（LEO）卫星系统因其能够实现快速和广域数据交换的能力而越来越受到重视，从而促进了跨地理分布区域的人工智能（AI）模型的协作培训。由于隐私问题和监管限制，在远程客户端收集的原始数据无法集中汇总，这对传统的人工智能训练方法构成了主要障碍。联邦学习通过在分布式设备上训练本地模型并仅交换模型参数，提供了一种保护隐私的替代方案。然而，卫星系统的动态拓扑结构和有限的带宽将阻碍及时的参数聚合和分发，导致训练时间延长。为了应对这一挑战，我们研究了在卫星网络上调度联邦学习的问题，并确定了影响每轮训练总体持续时间的关键瓶颈。我们提出了一个基于离散时间图的按需调度框架，动态分配通信资源，以加速联邦学习。仿真结果表明，所提出的方法实现了显着的性能增益比传统的基于统计复用的模型交换策略，减少了14.20%至41.48%的总轮时间。此外，对于更大的模型和更多的客户端，加速效果变得更加明显，突出了所提出的方法的可扩展性。
摘要：Large-scale low-Earth-orbit (LEO) satellite systems are increasingly valued for their ability to enable rapid and wide-area data exchange, thereby facilitating the collaborative training of artificial intelligence (AI) models across geographically distributed regions. Due to privacy concerns and regulatory constraints, raw data collected at remote clients cannot be centrally aggregated, posing a major obstacle to traditional AI training methods. Federated learning offers a privacy-preserving alternative by training local models on distributed devices and exchanging only model parameters. However, the dynamic topology and limited bandwidth of satellite systems will hinder timely parameter aggregation and distribution, resulting in prolonged training times. To address this challenge, we investigate the problem of scheduling federated learning over satellite networks and identify key bottlenecks that impact the overall duration of each training round. We propose a discrete temporal graph-based on-demand scheduling framework that dynamically allocates communication resources to accelerate federated learning. Simulation results demonstrate that the proposed approach achieves significant performance gains over traditional statistical multiplexing-based model exchange strategies, reducing overall round times by 14.20% to 41.48%. Moreover, the acceleration effect becomes more pronounced for larger models and higher numbers of clients, highlighting the scalability of the proposed approach.
推理|分析|理解|解释(7篇)
【1】ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement
标题：ChartGaze：通过眼动跟踪引导的注意力细化增强LVLM中的图表理解
链接：https://arxiv.org/abs/2509.13282
备注：EMNLP 2025
摘要：图表是沟通和表达信息的重要视觉媒介。虽然大型视觉语言模型（LVLM）在图表问题回答（CQA）方面取得了进展，但这项任务仍然具有挑战性，特别是当模型涉及图表的不相关区域时。在这项工作中，我们提出了ChartGaze，这是一个新的眼动跟踪数据集，可以在图表推理任务中捕获人类的凝视模式。通过对人类和模型注意力的系统比较，我们发现LVLM经常偏离人类视线，导致可解释性和准确性降低。为了解决这个问题，我们提出了一个凝视引导的注意力细化，使图像-文本注意力与人类的注视保持一致。我们的方法提高了答案的准确性和注意力的一致性，在多个模型中获得了高达2.56个百分点的收益。这些结果表明，结合人类的目光，以提高图表为重点的LVLM的推理质量和可解释性的承诺。
摘要 ：Charts are a crucial visual medium for communicating and representing information. While Large Vision-Language Models (LVLMs) have made progress on chart question answering (CQA), the task remains challenging, particularly when models attend to irrelevant regions of the chart. In this work, we present ChartGaze, a new eye-tracking dataset that captures human gaze patterns during chart reasoning tasks. Through a systematic comparison of human and model attention, we find that LVLMs often diverge from human gaze, leading to reduced interpretability and accuracy. To address this, we propose a gaze-guided attention refinement that aligns image-text attention with human fixations. Our approach improves both answer accuracy and attention alignment, yielding gains of up to 2.56 percentage points across multiple models. These results demonstrate the promise of incorporating human gaze to enhance both the reasoning quality and interpretability of chart-focused LVLMs.
【2】FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning
标题：FinSearchComp：对金融搜索和推理的现实，专家级评估
链接：https://arxiv.org/abs/2509.13160
备注：29 pages
摘要：搜索已成为基于LLM的代理的核心基础设施，并被广泛认为是迈向更通用智能的关键。金融是一个特别苛刻的试验场：分析师经常对时间敏感的特定领域数据进行复杂的多步搜索，这使得它成为评估搜索熟练度和基于知识的推理的理想选择。然而，目前还没有开放的金融数据集评估端到端代理的数据搜索能力，这主要是因为构建现实的、复杂的任务需要深厚的金融专业知识，而时间敏感的数据很难评估。我们提出FinSearchComp，第一个完全开源的代理基准，用于现实的，开放领域的金融搜索和推理。FinSearchComp包括三个任务-时间敏感数据获取，简单历史查询和复杂历史调查-紧密再现真实世界的金融分析师工作流程。为了确保难度和可靠性，我们聘请了70位专业的金融专家进行注释，并实施了严格的多阶段质量保证管道。该基准测试包括635个问题，涵盖全球和大中华区市场，我们评估了21个模型（产品）。Grok 4（网络）在全球子集中名列前茅，接近专家级的准确性。DouBao（网络）在大中华区领先。实验分析表明，为代理人配备网络搜索和金融插件可以大大改善FinSearchComp的结果，模型和工具的国家来源对性能有显著影响。通过与实际分析任务保持一致并提供端到端评估，FinSearchComp为复杂的金融搜索和推理提供了一个专业的，高难度的测试平台。
摘要：Search has emerged as core infrastructure for LLM-based agents and is widely viewed as critical on the path toward more general intelligence. Finance is a particularly demanding proving ground: analysts routinely conduct complex, multi-step searches over time-sensitive, domain-specific data, making it ideal for assessing both search proficiency and knowledge-grounded reasoning. Yet no existing open financial datasets evaluate data searching capability of end-to-end agents, largely because constructing realistic, complicated tasks requires deep financial expertise and time-sensitive data is hard to evaluate. We present FinSearchComp, the first fully open-source agent benchmark for realistic, open-domain financial search and reasoning. FinSearchComp comprises three tasks -- Time-Sensitive Data Fetching, Simple Historical Lookup, and Complex Historical Investigation -- closely reproduce real-world financial analyst workflows. To ensure difficulty and reliability, we engage 70 professional financial experts for annotation and implement a rigorous multi-stage quality-assurance pipeline. The benchmark includes 635 questions spanning global and Greater China markets, and we evaluate 21 models (products) on it. Grok 4 (web) tops the global subset, approaching expert-level accuracy. DouBao (web) leads on the Greater China subset. Experimental analyses show that equipping agents with web search and financial plugins substantially improves results on FinSearchComp, and the country origin of models and tools impact performance significantly.By aligning with realistic analyst tasks and providing end-to-end evaluation, FinSearchComp offers a professional, high-difficulty testbed for complex financial search and reasoning.
【3】Comparative Analysis of Wave Scattering Numerical Modeling Using the Boundary Element Method and Physics-Informed Neural Networks
标题：边界元法和物理神经网络波浪散射数值模拟的比较分析
链接：https://arxiv.org/abs/2509.12483
备注：19 pages, 7 figures
摘要：目的-本研究比较了边界元法（BEM）和物理信息神经网络（PINNs）求解二维亥姆霍兹方程的波散射问题。目的是在相同条件下评价两种方法的性能。 设计/方法/方式-我们解决了亥姆霍兹方程使用边界元法和PINNs相同的散射问题。PINN通过最小化控制方程和边界条件的残差来训练，其配置通过超参数优化确定，而BEM则使用边界离散化来应用。这两种方法进行评估的解决方案的精度，计算时间和泛化能力。 结果-数值实验进行了不同的整合点的数量为BEM和PINN的层数和每层的神经元。超参数调整提供了进一步的洞察合适的配置波散射问题。在相当的精度下，PINN产生了一致的解决方案，但需要的训练时间比BEM长约42倍。然而，一旦经过训练，PINN的评估时间快了204倍。在PINN训练域之外也评估了泛化能力，其中相对误差从7.46 × 10^{-2}$增加到8.22，而BEM在扩展区域中保持类似的误差水平。 独创性/价值-这项工作提出了一个直接比较PINNs和边界元的亥姆霍兹方程。分析提供了定量数据的性能，这两种方法，支持他们的选择在未来的研究波传播问题，并建立未来的挑战和方向。
摘要：Purpose - This study compares the Boundary Element Method (BEM) and Physics-Informed Neural Networks (PINNs) for solving the two-dimensional Helmholtz equation in wave scattering problems. The objective is to evaluate the performance of both methods under the same conditions. Design/methodology/approach - We solve the Helmholtz equation using BEM and PINNs for the same scattering problem. The PINNs are trained by minimizing the residual of the governing equations and boundary conditions, with their configuration determined through hyperparameter optimization, while the BEM is applied using boundary discretization. Both methods are evaluated in terms of solution accuracy, computation time, and generalization capacity. Findings - Numerical experiments were conducted by varying the number of integration points for BEM and the number of layers and neurons per layer for PINNs. Hyperparameter tuning provided further insight into suitable configurations for wave scattering problems. At comparable accuracy, PINNs produced consistent solutions but required training times approximately 42 times longer than BEM. However, once trained, PINNs achieved evaluation times up to 204 times faster. The generalization capacity was also assessed outside the PINN training domain, where the relative error increased from $7.46 \times 10^{-2}$ to 8.22, while BEM maintained a similar error level in the extended region. Originality/value - This work presents a direct comparison between PINNs and BEM for the Helmholtz equation. The analysis provides quantitative data on the performance of both methods, supporting their selection in future research on wave propagation problems and establishing future challenges and directions.
【4】Surrogate Representation Inference for Noisy Text and Image Annotations
标题：嘈杂文本和图像注释的替代表示推理
链接：https://arxiv.org/abs/2509.12416
摘要：随着研究人员越来越多地依赖机器学习模型和LLM来注释非结构化数据，如文本或图像，已经提出了各种方法来纠正下游统计分析中的偏差。然而，现有的方法往往会产生较大的标准误差，并需要一些无错误的人工注释。在本文中，我介绍了代理表示推理（SRI），它假设非结构化数据完全调解人类注释和结构化变量之间的关系。假设人类编码人员仅依赖于非结构化数据进行注释，则设计保证了这一假设。在这种情况下，我提出了一个神经网络架构，学习非结构化数据的低维表示，使代理假设仍然是满足的。当多个人类注释可用时，SRI可以进一步校正可能存在于人类注释中的非差分测量误差。专注于文本作为结果的设置，我正式建立的识别条件和半参数有效的估计策略，使学习和利用这样的低维表示。仿真研究和实际应用表明，当机器学习预测精度适中时，SRI将标准误差降低了50%以上，即使人类注释包含非差分测量误差，也能提供有效的推理。
摘要 ：As researchers increasingly rely on machine learning models and LLMs to annotate unstructured data, such as texts or images, various approaches have been proposed to correct bias in downstream statistical analysis. However, existing methods tend to yield large standard errors and require some error-free human annotation. In this paper, I introduce Surrogate Representation Inference (SRI), which assumes that unstructured data fully mediate the relationship between human annotations and structured variables. The assumption is guaranteed by design provided that human coders rely only on unstructured data for annotation. Under this setting, I propose a neural network architecture that learns a low-dimensional representation of unstructured data such that the surrogate assumption remains to be satisfied. When multiple human annotations are available, SRI can further correct non-differential measurement errors that may exist in human annotations. Focusing on text-as-outcome settings, I formally establish the identification conditions and semiparametric efficient estimation strategies that enable learning and leveraging such a low-dimensional representation. Simulation studies and a real-world application demonstrate that SRI reduces standard errors by over 50% when machine learning prediction accuracy is moderate and provides valid inference even when human annotations contain non-differential measurement errors.
【5】InPhyRe Discovers: Large Multimodal Models Struggle in Inductive Physical Reasoning
标题：InPhyRe发现：大型多模态模型在归纳物理推理中挣扎
链接：https://arxiv.org/abs/2509.12263
备注：35 pages including appendix
摘要：大型多模态模型（LVMs）将训练过程中观察到的普遍物理定律（如动量守恒）编码为参数知识。它允许Lempo回答物理推理查询，例如来自视觉输入的潜在碰撞事件的结果。然而，由于参数知识只包括在训练过程中看到的物理定律，因此当推理场景违反这些物理定律时，它不足以进行推理。相比之下，人类拥有的技能，以适应他们的物理推理看不见的物理环境，从一些视觉的例子。这种能力，我们称之为归纳物理推理，是必不可少的，如果他们要取代人类代理在安全关键的应用程序。尽管它的重要性，现有的视觉基准评估只有参数的知识，而不是归纳的物理推理。为此，我们提出了InPhyRe，第一个视觉问答基准，以衡量归纳物理推理在Lancaster。InPhyRe评估了Lyndrome在算法生成的合成碰撞视频中预测碰撞事件结果的能力。通过检查13个Lyndrome，InPhyRe告诉我们：（1）Lyndrome努力将其关于普遍物理定律的有限参数知识应用于推理，（2）当演示样本违反普遍物理定律时，Lyndrome中的归纳物理推理很弱，（3）Lyndrome中的归纳物理推理存在语言偏见，并且在很大程度上忽略了视觉输入，质疑Lyndrome关于视觉输入的可信度。
摘要：Large multimodal models (LMMs) encode universal physical laws observed during training, such as momentum conservation, as parametric knowledge. It allows LMMs to answer physical reasoning queries, such as the outcome of a potential collision event from visual input. However, since parametric knowledge includes only the physical laws seen during training, it is insufficient for reasoning when the inference scenario violates these physical laws. In contrast, humans possess the skill to adapt their physical reasoning to unseen physical environments from a few visual examples. This ability, which we refer to as inductive physical reasoning, is indispensable for LMMs if they are to replace human agents in safety-critical applications. Despite its importance, existing visual benchmarks evaluate only the parametric knowledge in LMMs, and not inductive physical reasoning. To this end, we propose InPhyRe, the first visual question answering benchmark to measure inductive physical reasoning in LMMs. InPhyRe evaluates LMMs on their ability to predict the outcome of collision events in algorithmically generated synthetic collision videos. By inspecting 13 LMMs, InPhyRe informs us that (1) LMMs struggle to apply their limited parametric knowledge about universal physical laws to reasoning, (2) inductive physical reasoning in LMMs is weak when demonstration samples violate universal physical laws, and (3) inductive physical reasoning in LMMs suffers from language bias and largely ignores the visual inputs, questioning the trustworthiness of LMMs regarding visual inputs.
【6】Explainable Fraud Detection with GNNExplainer and Shapley Values
标题：基于GNNExplainer和Shapley值的可解释欺诈检测
链接：https://arxiv.org/abs/2509.12262
备注：B. Comp Dissertation
摘要：随着数字支付的使用越来越频繁，金融欺诈的风险也在增加。虽然人工智能系统在欺诈检测中的应用非常广泛，但社会和监管机构已经提高了这些系统的透明度标准，以实现可靠性验证。为了提高他们在进行欺诈调查的有效性，欺诈分析师还受益于简明易懂的解释。为了解决这些挑战，本文将集中精力开发一个可解释的欺诈检测器。
摘要：The risk of financial fraud is increasing as digital payments are used more and more frequently. Although the use of artificial intelligence systems for fraud detection is widespread, society and regulators have raised the standards for these systems' transparency for reliability verification purposes. To increase their effectiveness in conducting fraud investigations, fraud analysts also profit from having concise and understandable explanations. To solve these challenges, the paper will concentrate on developing an explainable fraud detector.
【7】Towards Trustworthy Agentic IoEV: AI Agents for Explainable Cyberthreat Mitigation and State Analytics
标题：迈向值得信赖的大型IoEV：用于可解释的网络威胁缓解和状态分析的人工智能代理
链接：https://arxiv.org/abs/2509.12233
备注：10 pages, 7 figures, Accepted at LCN'25
摘要：电动汽车互联网（IoEV）设想了一个由电动汽车（EV）、充电基础设施和电网服务紧密耦合的生态系统，但它仍然容易受到网络攻击、不可靠的电池状态预测和不透明的决策过程的影响，这些都会削弱信任和性能。为了应对这些挑战，我们引入了一种为IoEV量身定制的新型人工智能（AAI）框架，其中专门的代理合作提供自主威胁缓解，强大的分析和可解释的决策支持。具体来说，我们设计了一个AAI架构，包括专用代理，用于充电站的网络威胁检测和响应，实时充电状态（SoC）估计和健康状态（SoH）异常检测，所有这些都通过共享的可解释推理层进行协调;开发可解释的威胁缓解机制，主动识别和中和对物理充电点和学习组件的攻击;提出弹性SoC和SoH模型，利用持续和对抗感知学习来产生准确的，具有人类可读解释的不确定性感知预测;并实现三代理管道，其中每个代理使用LLM驱动的推理和动态工具调用来解释意图，将任务置于上下文环境中，并执行以用户为中心的辅助的正式优化。最后，我们通过在不同的IoEV场景中进行全面的实验来验证我们的框架，证明了安全性和预测准确性的显着改进。所有数据集、模型和代码都将公开发布。
摘要：The Internet of Electric Vehicles (IoEV) envisions a tightly coupled ecosystem of electric vehicles (EVs), charging infrastructure, and grid services, yet it remains vulnerable to cyberattacks, unreliable battery-state predictions, and opaque decision processes that erode trust and performance. To address these challenges, we introduce a novel Agentic Artificial Intelligence (AAI) framework tailored for IoEV, where specialized agents collaborate to deliver autonomous threat mitigation, robust analytics, and interpretable decision support. Specifically, we design an AAI architecture comprising dedicated agents for cyber-threat detection and response at charging stations, real-time State of Charge (SoC) estimation, and State of Health (SoH) anomaly detection, all coordinated through a shared, explainable reasoning layer; develop interpretable threat-mitigation mechanisms that proactively identify and neutralize attacks on both physical charging points and learning components; propose resilient SoC and SoH models that leverage continuous and adversarial-aware learning to produce accurate, uncertainty-aware forecasts with human-readable explanations; and implement a three-agent pipeline, where each agent uses LLM-driven reasoning and dynamic tool invocation to interpret intent, contextualize tasks, and execute formal optimizations for user-centric assistance. Finally, we validate our framework through comprehensive experiments across diverse IoEV scenarios, demonstrating significant improvements in security and prediction accuracy. All datasets, models, and code will be released publicly.
检测相关(4篇)
【1】Dual-Stage Reweighted MoE for Long-Tailed Egocentric Mistake Detection
标题：双阶段重新加权MoE用于长尾自我中心错误检测
链接：https://arxiv.org/abs/2509.12990
摘要：在这份报告中，我们解决的问题，确定用户是否执行一个动作不正确地从以自我为中心的视频数据。为了应对微妙和罕见的错误所带来的挑战，我们提出了一个双阶段重新加权混合专家（DR-MoE）框架。在第一阶段中，使用冻结的ViViT模型和LoRA调整的ViViT模型提取特征，这些模型通过特征级专家模块进行组合。在第二阶段，三个分类器被训练有不同的目标：重新加权的交叉熵，以减轻类的不平衡，AUC损失，以改善在偏态分布下的排名，和标签感知损失与清晰度感知最小化，以增强校准和泛化。他们的预测融合使用分类级专家模块。所提出的方法实现了强大的性能，特别是在识别罕见的和模糊的错误实例。该代码可在https://github.com/boyuh/DR-MoE上获得。
摘要：In this report, we address the problem of determining whether a user performs an action incorrectly from egocentric video data. To handle the challenges posed by subtle and infrequent mistakes, we propose a Dual-Stage Reweighted Mixture-of-Experts (DR-MoE) framework. In the first stage, features are extracted using a frozen ViViT model and a LoRA-tuned ViViT model, which are combined through a feature-level expert module. In the second stage, three classifiers are trained with different objectives: reweighted cross-entropy to mitigate class imbalance, AUC loss to improve ranking under skewed distributions, and label-aware loss with sharpness-aware minimization to enhance calibration and generalization. Their predictions are fused using a classification-level expert module. The proposed method achieves strong performance, particularly in identifying rare and ambiguous mistake instances. The code is available at https://github.com/boyuh/DR-MoE.
【2】Leveraging Intermediate Representations of Time Series Foundation Models for Anomaly Detection
标题：利用时间序列基础模型的中间表示进行异常检测
链接：https://arxiv.org/abs/2509.12650
备注：10 pages,8 figures
摘要：检测时间序列数据中的异常对于许多现实世界系统的可靠运行至关重要。最近，时间序列基础模型（TSFMs）已成为一个强大的工具，异常检测。然而，现有的方法通常依赖于TSFM的最后一层的表示，通过特定于任务的头将异常分数计算为重建或预测误差。相反，我们提出了TimeRep，一种新的异常检测方法，利用中间层的TSFM表示，计算这些表示之间的距离的异常分数。给定一个预先训练的TSFM，TimeRep选择中间层和补丁令牌位置，以产生最具信息性的表示。TimeRep从训练数据中形成中间表示的参考集合，并应用核心集策略来减小其大小，同时保持分布覆盖。在推理过程中，TimeRep通过测量传入数据的中间表示与集合表示之间的距离来计算传入数据的异常分数。为了解决概念漂移，TimeRep集成了一个自适应机制，在推理时，专门使用来自传入数据的非冗余中间表示来增强集合。我们对UCR Anomaly Archive进行了广泛的实验，其中包含250个单变量时间序列。TimeRep始终优于广泛的最先进的基线，包括非DL，DL和基于基础模型的方法。
摘要：Detecting anomalies in time series data is essential for the reliable operation of many real-world systems. Recently, time series foundation models (TSFMs) have emerged as a powerful tool for anomaly detection. However, existing methods typically rely on the final layer's representations of TSFMs, computing the anomaly score as a reconstruction or forecasting error via a task-specific head. Instead, we propose TimeRep, a novel anomaly detection approach that leverages the intermediate layer's representations of TSFMs, computing the anomaly score as the distance between these representations. Given a pre-trained TSFM, TimeRep selects the intermediate layer and patch-token position that yield the most informative representation. TimeRep forms a reference collection of intermediate representations from the training data and applies a core-set strategy to reduce its size while maintaining distributional coverage. During inference, TimeRep computes the anomaly score for incoming data by measuring the distance between its intermediate representations and those of the collection. To address concept drift, TimeRep integrates an adaptation mechanism that, at inference time, augments the collection exclusively with non-redundant intermediate representations from incoming data. We conducted extensive experiments on the UCR Anomaly Archive, which contains 250 univariate time series. TimeRep consistently outperforms a broad spectrum of state-of-the-art baselines, including non-DL, DL, and foundation model-based methods.
【3】Cross-Modal Deep Metric Learning for Time Series Anomaly Detection
标题：用于时间序列异常检测的跨模式深度度量学习
链接：https://arxiv.org/abs/2509.12540
摘要：为有效解决时间序列异常检测灵敏度低、耗时长的问题，提出一种基于跨模态深度度量学习的异常检测方法。构建了跨模态深度度量学习特征聚类模型，该模型由输入层、三元组选择层和损失函数计算层组成。计算聚类中心之间的平方欧几里德距离，并采用随机梯度下降策略来优化模型和分类不同的时间序列特征。主分量方向向量的内积被用作异常测量的度量。采用von Mises-Fisher（vMF）分布描述时间序列数据的方向性特征，利用历史数据进行训练，得到评价参数。通过比较实际时间序列数据的主成分方向向量与阈值，进行异常检测。实验结果表明，该方法能够准确地对不同属性的时间序列数据进行分类，对异常具有较高的敏感性，检测精度高、检测速度快、鲁棒性强。
摘要：To effectively address the issues of low sensitivity and high time consumption in time series anomaly detection, we propose an anomaly detection method based on cross-modal deep metric learning. A cross-modal deep metric learning feature clustering model is constructed, composed of an input layer, a triplet selection layer, and a loss function computation layer. The squared Euclidean distances between cluster centers are calculated, and a stochastic gradient descent strategy is employed to optimize the model and classify different time series features. The inner product of principal component direction vectors is used as a metric for anomaly measurement. The von Mises-Fisher (vMF) distribution is applied to describe the directional characteristics of time series data, and historical data is used to train and obtain evaluation parameters. By comparing the principal component direction vector of actual time series data with the threshold, anomaly detection is performed. Experimental results demonstrate that the proposed method accurately classifies time series data with different attributes, exhibits high sensitivity to anomalies, and achieves high detection accuracy, fast detection speed, and strong robustness.
【4】Cott-ADNet: Lightweight Real-Time Cotton Boll and Flower Detection Under Field Conditions
标题：Cott-ADNet：田间条件下轻量级实时棉花和花朵检测
链接：https://arxiv.org/abs/2509.12442
备注：14 pages, 5 figures, 1 table
摘要：棉花是世界上最重要的天然纤维作物之一，但收获仍然受到劳动密集型人工采摘、低效率和错过最佳收获窗口造成的产量损失的限制。因此，准确识别棉铃及其成熟度对于自动化、产量估计和育种研究至关重要。我们提出了棉花ADNet，一个轻量级的实时检测器，适合复杂的田间条件下的棉铃和花朵识别。基于YOLOv 11n，Cott-ADNet通过改进的卷积设计增强了空间表示和鲁棒性，同时引入了两个新模块：NeLU增强的全局注意力机制，以更好地捕获弱和低对比度特征，以及扩大的感受野SPPF，以低计算成本扩展感受野，实现更有效的多尺度上下文建模。我们策划了一个包含4，966张图像的标记数据集，并发布了一个包含1，216张现场图像的外部验证集，以支持未来的研究。实验结果表明，Cott-ADNet仅用7.5 GFLOPs就实现了91.5%的准确率、89.8%的召回率、93.3%的mAP 50、71.3%的mAP和90.6%的F1-Score，在多尺度和旋转变化下保持稳定的性能。这些结果表明，Cott-ADNet是一种准确有效的田间部署解决方案，从而为自动化棉花收获和高通量表型分析提供了可靠的基础。代码和数据集可在https://github.com/SweefongWong/Cott-ADNet上获得。
摘要 ：Cotton is one of the most important natural fiber crops worldwide, yet harvesting remains limited by labor-intensive manual picking, low efficiency, and yield losses from missing the optimal harvest window. Accurate recognition of cotton bolls and their maturity is therefore essential for automation, yield estimation, and breeding research. We propose Cott-ADNet, a lightweight real-time detector tailored to cotton boll and flower recognition under complex field conditions. Building on YOLOv11n, Cott-ADNet enhances spatial representation and robustness through improved convolutional designs, while introducing two new modules: a NeLU-enhanced Global Attention Mechanism to better capture weak and low-contrast features, and a Dilated Receptive Field SPPF to expand receptive fields for more effective multi-scale context modeling at low computational cost. We curate a labeled dataset of 4,966 images, and release an external validation set of 1,216 field images to support future research. Experiments show that Cott-ADNet achieves 91.5% Precision, 89.8% Recall, 93.3% mAP50, 71.3% mAP, and 90.6% F1-Score with only 7.5 GFLOPs, maintaining stable performance under multi-scale and rotational variations. These results demonstrate Cott-ADNet as an accurate and efficient solution for in-field deployment, and thus provide a reliable basis for automated cotton harvesting and high-throughput phenotypic analysis. Code and dataset is available at https://github.com/SweefongWong/Cott-ADNet.
分类|识别(5篇)
【1】TimeCluster with PCA is Equivalent to Subspace Identification of Linear Dynamical Systems
标题：带PCA的Time集群等效于线性动态系统的子空间识别
链接：https://arxiv.org/abs/2509.12895
备注：15 pages, 9 figures
摘要：TimeCluster是一种可视化分析技术，通过将重叠的数据窗口投影到低维空间中来发现长多元时间序列中的结构。我们表明，当主成分分析（PCA）被选为降维技术，这个过程在数学上相当于经典的线性子空间识别（块汉克尔矩阵加奇异向量分解（SVD））。在这两种方法中，从时间序列数据中提取相同的低维线性子空间。本文首先回顾了TimeCluster方法和子空间系统辨识理论。然后，我们表明，形成的时间序列的滑动窗口矩阵产生一个汉克尔矩阵，所以应用PCA（通过SVD）这个矩阵恢复相同的主方向子空间识别。因此，来自TimeCluster的聚类坐标与子空间识别方法一致。我们目前的实验合成和真正的动态信号确认这两个嵌入相吻合。最后，我们探讨和讨论了这种等价性带来的未来机会，包括从确定的状态空间进行预测，流/在线扩展，整合和可视化外部输入以及显示损坏数据潜在趋势的强大技术。
摘要：TimeCluster is a visual analytics technique for discovering structure in long multivariate time series by projecting overlapping windows of data into a low-dimensional space. We show that, when Principal Component Analysis (PCA) is chosen as the dimensionality reduction technique, this procedure is mathematically equivalent to classical linear subspace identification (block-Hankel matrix plus Singular Vector Decomposition (SVD)). In both approaches, the same low-dimensional linear subspace is extracted from the time series data. We first review the TimeCluster method and the theory of subspace system identification. Then we show that forming the sliding-window matrix of a time series yields a Hankel matrix, so applying PCA (via SVD) to this matrix recovers the same principal directions as subspace identification. Thus the cluster coordinates from TimeCluster coincide with the subspace identification methods. We present experiments on synthetic and real dynamical signals confirming that the two embeddings coincide. Finally, we explore and discuss future opportunities enabled by this equivalence, including forecasting from the identified state space, streaming/online extensions, incorporating and visualising external inputs and robust techniques for displaying underlying trends in corrupted data.
【2】ZTree: A Subgroup Identification Based Decision Tree Learning Framework
标题：ZTree：一个基于子组识别的决策树学习框架
链接：https://arxiv.org/abs/2509.12688
备注：15 pages, 1 table, 5 figures
摘要：决策树是一种常用的机器学习模型，因其可解释性和多功能性而受到重视，能够进行分类和回归。我们提出了ZTree，一个新的决策树学习框架，取代CART的传统纯度为基础的分裂与统计原则的子群识别。在每个节点处，ZTree应用假设检验（例如，z-检验、t-检验、Mann-Whitney U、对数秩），以评估候选亚组是否与补体有意义地不同。为了适应多重测试的复杂性，我们采用了基于交叉验证的方法来确定是否需要进一步的节点分裂。这种鲁棒的停止准则消除了对后修剪的需要，并且使得测试阈值（z阈值）成为用于控制树复杂性的唯一参数。由于树生长过程的简单性，一旦使用最宽松的z阈值学习了详细的树，则可以通过简单地移除不满足较大z阈值的节点来导出所有更简单的树。这使得参数调整直观而高效。此外，这个z阈值本质上是一个p值，允许用户轻松地将适当的统计测试插入我们的框架，而无需调整参数搜索的范围。对五个大规模UCI数据集的实证评估表明，ZTree始终提供强大的性能，特别是在低数据状态下。与CART相比，ZTree也倾向于在不牺牲性能的情况下生长更简单的树。ZTree通过利用假设检验和交叉验证方法来进行多重检验校正，引入了一种基于统计的替代传统决策树分裂的方法，从而形成了一个高效灵活的框架。
摘要：Decision trees are a commonly used class of machine learning models valued for their interpretability and versatility, capable of both classification and regression. We propose ZTree, a novel decision tree learning framework that replaces CART's traditional purity based splitting with statistically principled subgroup identification. At each node, ZTree applies hypothesis testing (e.g., z-tests, t-tests, Mann-Whitney U, log-rank) to assess whether a candidate subgroup differs meaningfully from the complement. To adjust for the complication of multiple testing, we employ a cross-validation-based approach to determine if further node splitting is needed. This robust stopping criterion eliminates the need for post-pruning and makes the test threshold (z-threshold) the only parameter for controlling tree complexity. Because of the simplicity of the tree growing procedure, once a detailed tree is learned using the most lenient z-threshold, all simpler trees can be derived by simply removing nodes that do not meet the larger z-thresholds. This makes parameter tuning intuitive and efficient. Furthermore, this z-threshold is essentially a p-value, allowing users to easily plug in appropriate statistical tests into our framework without adjusting the range of parameter search. Empirical evaluation on five large-scale UCI datasets demonstrates that ZTree consistently delivers strong performance, especially at low data regimes. Compared to CART, ZTree also tends to grow simpler trees without sacrificing performance. ZTree introduces a statistically grounded alternative to traditional decision tree splitting by leveraging hypothesis testing and a cross-validation approach to multiple testing correction, resulting in an efficient and flexible framework.
【3】Linear Dimensionality Reduction for Word Embeddings in Tabular Data Classification
标题：表格数据分类中单词嵌入的线性简化
链接：https://arxiv.org/abs/2509.12346
摘要：工程师的工资预测挑战要求根据表格数据将工资类别分为三类。职位描述被表示为一个300维的单词嵌入到表格特征中，大大增加了维度。此外，有限数量的训练样本使得分类具有挑战性。用于表格数据分类的词嵌入的线性降维仍然未被探索。本文研究了主成分分析（PCA）和线性判别分析（LDA）。我们表明，PCA，与适当的子空间维数，可以优于原始嵌入。由于协方差估计误差，没有正则化的LDA表现不佳，但应用收缩显著提高了性能，即使只有两个维度。我们提出了分区LDA，它将嵌入分割成大小相等的块，并分别对每个块执行LDA，从而减少协方差矩阵的大小。分区LDA优于常规LDA，并结合收缩，在竞争公共排行榜上达到前10名的准确性。该方法在训练样本有限的表格数据分类中有效地提高了词嵌入性能。
摘要：The Engineers' Salary Prediction Challenge requires classifying salary categories into three classes based on tabular data. The job description is represented as a 300-dimensional word embedding incorporated into the tabular features, drastically increasing dimensionality. Additionally, the limited number of training samples makes classification challenging. Linear dimensionality reduction of word embeddings for tabular data classification remains underexplored. This paper studies Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). We show that PCA, with an appropriate subspace dimension, can outperform raw embeddings. LDA without regularization performs poorly due to covariance estimation errors, but applying shrinkage improves performance significantly, even with only two dimensions. We propose Partitioned-LDA, which splits embeddings into equal-sized blocks and performs LDA separately on each, thereby reducing the size of the covariance matrices. Partitioned-LDA outperforms regular LDA and, combined with shrinkage, achieves top-10 accuracy on the competition public leaderboard. This method effectively enhances word embedding performance in tabular data classification with limited training samples.
【4】More Similar than Dissimilar: Modeling Annotators for Cross-Corpus Speech Emotion Recognition
标题：相似多于不同：跨数据库语音情感识别的注释器建模
链接：https://arxiv.org/abs/2509.12295
备注：©20XX IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
摘要：语音情感识别系统通常预测从多个注释者的评级生成的共识值。然而，这些模型预测任何一个人的注释的能力有限。或者，模型可以学习预测所有注释器的注释。使这些模型适应新的注释器是困难的，因为新的注释器必须单独提供足够的标记训练数据。我们建议通过使用在大型注释者群体上预训练的模型来利用注释者之间的相似性，以识别类似的，以前见过的注释者。给定一个新的、以前看不见的注释器和有限的注册数据，我们可以对类似的注释器进行预测，从而对目标数据集中看不见的数据进行现成的注释，从而提供一种成本极低的个性化机制。我们证明了我们的方法显着优于其他现成的方法，为轻量级的情感适应，实际的现实世界的部署铺平了道路。
摘要：Speech emotion recognition systems often predict a consensus value generated from the ratings of multiple annotators. However, these models have limited ability to predict the annotation of any one person. Alternatively, models can learn to predict the annotations of all annotators. Adapting such models to new annotators is difficult as new annotators must individually provide sufficient labeled training data. We propose to leverage inter-annotator similarity by using a model pre-trained on a large annotator population to identify a similar, previously seen annotator. Given a new, previously unseen, annotator and limited enrollment data, we can make predictions for a similar annotator, enabling off-the-shelf annotation of unseen data in target datasets, providing a mechanism for extremely low-cost personalization. We demonstrate our approach significantly outperforms other off-the-shelf approaches, paving the way for lightweight emotion adaptation, practical for real-world deployment.
【5】Flow-Based Fragment Identification via Binding Site-Specific Latent Representations
标题：通过结合位点特定潜在代表进行基于流的片段识别
链接：https://arxiv.org/abs/2509.13216
摘要：基于片段的药物设计是一种很有前途的策略，利用小化学部分的结合，可以有效地指导药物发现。片段鉴定的初始步骤仍然具有挑战性，因为片段通常结合较弱且非特异性。我们开发了一种蛋白质片段编码器，它依赖于对比学习方法来映射共享潜在空间中的分子片段和蛋白质表面。编码器捕获交互相关的功能，并允许执行虚拟筛选以及生成设计与我们的新方法LatentFrag。在LatentFrag中，片段嵌入和位置是在蛋白质表面条件下生成的，同时通过构建而具有化学现实性。我们的表达片段和蛋白质表示允许定位蛋白质-片段相互作用位点，具有高灵敏度，并且当从学习的潜在片段嵌入分布中采样时，我们观察到最先进的片段恢复率。我们的生成方法优于常见的方法，如虚拟筛选在其计算成本的一小部分，提供了一个有价值的起点片段命中发现。我们进一步展示了LatentFrag的实用性，并将工作流程扩展到完整的配体设计任务。总之，这些方法有助于推进片段鉴定，并为基于片段的药物发现提供有价值的工具。
摘要：Fragment-based drug design is a promising strategy leveraging the binding of small chemical moieties that can efficiently guide drug discovery. The initial step of fragment identification remains challenging, as fragments often bind weakly and non-specifically. We developed a protein-fragment encoder that relies on a contrastive learning approach to map both molecular fragments and protein surfaces in a shared latent space. The encoder captures interaction-relevant features and allows to perform virtual screening as well as generative design with our new method LatentFrag. In LatentFrag, fragment embeddings and positions are generated conditioned on the protein surface while being chemically realistic by construction. Our expressive fragment and protein representations allow location of protein-fragment interaction sites with high sensitivity and we observe state-of-the-art fragment recovery rates when sampling from the learned distribution of latent fragment embeddings. Our generative method outperforms common methods such as virtual screening at a fraction of its computational cost providing a valuable starting point for fragment hit discovery. We further show the practical utility of LatentFrag and extend the workflow to full ligand design tasks. Together, these approaches contribute to advancing fragment identification and provide valuable tools for fragment-based drug discovery.
表征(3篇)
【1】Improving Accuracy and Efficiency of Implicit Neural Representations: Making SIREN a WINNER
标题：提高隐式神经表示的准确性和效率：让SIREN成为赢家
链接：https://arxiv.org/abs/2509.12980
摘要：我们确定并解决了正弦表示网络（SIREN），一类隐式神经表示的基本限制。SIRENs Sitzmann et al.（2020），当没有适当初始化时，可能会在拟合超出其频率支持范围的信号时遇到困难。在极端情况下，当网络的频率支持与目标频谱不一致时，会观察到“频谱瓶颈”现象，其中模型产生接近零的输出，甚至无法恢复其代表能力内的频率分量。为了克服这个问题，我们提出了WINNER -带噪声的神经表示权重初始化。WINNER用高斯噪声扰动基本SIREN的均匀初始化权重-其噪声尺度由目标信号的频谱质心自适应地确定。类似于随机傅立叶嵌入，这减轻了“频谱偏差”，但不引入额外的可训练参数。我们的方法实现了最先进的音频拟合和显着的收益，在图像和3D形状拟合任务的基础SIREN。除了信号拟合之外，WINNER还提出了自适应目标感知初始化策略的新途径，以优化深度神经网络训练。代码和数据请访问cfdlabtechnion.github.io/siren_square/。
摘要：We identify and address a fundamental limitation of sinusoidal representation networks (SIRENs), a class of implicit neural representations. SIRENs Sitzmann et al. (2020), when not initialized appropriately, can struggle at fitting signals that fall outside their frequency support. In extreme cases, when the network's frequency support misaligns with the target spectrum, a 'spectral bottleneck' phenomenon is observed, where the model yields to a near-zero output and fails to recover even the frequency components that are within its representational capacity. To overcome this, we propose WINNER - Weight Initialization with Noise for Neural Representations. WINNER perturbs uniformly initialized weights of base SIREN with Gaussian noise - whose noise scales are adaptively determined by the spectral centroid of the target signal. Similar to random Fourier embeddings, this mitigates 'spectral bias' but without introducing additional trainable parameters. Our method achieves state-of-the-art audio fitting and significant gains in image and 3D shape fitting tasks over base SIREN. Beyond signal fitting, WINNER suggests new avenues in adaptive, target-aware initialization strategies for optimizing deep neural network training. For code and data visit cfdlabtechnion.github.io/siren_square/.
【2】Representation Learning on Large Non-Bipartite Transaction Networks using GraphSAGE
标题：使用GraphSAGE进行大型非双方交易网络的表示学习
链接：https://arxiv.org/abs/2509.12255
备注：None
摘要 ：金融机构越来越需要可扩展的工具来分析复杂的交易网络，但传统的图嵌入方法难以处理动态的真实银行数据。本文展示了GraphSAGE的实际应用，一个归纳图神经网络框架，在银行的背景下，非二分异构交易网络。与转换方法不同，GraphSAGE可以很好地扩展到大型网络，并且可以推广到看不见的节点，这对于处理随时间变化的交易数据的机构来说至关重要。我们构建了一个交易网络，使用匿名的客户和商家的交易和训练GraphSAGE模型生成节点嵌入。我们对嵌入的探索性工作揭示了与地理和人口属性一致的可解释集群。此外，我们通过将它们应用于钱骡检测模型来说明它们在下游分类任务中的实用性，在该模型中使用这些嵌入提高了高风险账户的优先级。除了欺诈检测之外，我们的工作还强调了该框架对银行规模网络的适应性，强调了其归纳能力、可扩展性和可解释性。这项研究为金融组织提供了一个蓝图，利用图形机器学习在交易生态系统中获得可操作的见解。
摘要：Financial institutions increasingly require scalable tools to analyse complex transactional networks, yet traditional graph embedding methods struggle with dynamic, real-world banking data. This paper demonstrates the practical application of GraphSAGE, an inductive Graph Neural Network framework, to non-bipartite heterogeneous transaction networks within a banking context. Unlike transductive approaches, GraphSAGE scales well to large networks and can generalise to unseen nodes which is critical for institutions working with temporally evolving transactional data. We construct a transaction network using anonymised customer and merchant transactions and train a GraphSAGE model to generate node embeddings. Our exploratory work on the embeddings reveals interpretable clusters aligned with geographic and demographic attributes. Additionally, we illustrate their utility in downstream classification tasks by applying them to a money mule detection model where using these embeddings improves the prioritisation of high-risk accounts. Beyond fraud detection, our work highlights the adaptability of this framework to banking-scale networks, emphasising its inductive capability, scalability, and interpretability. This study provides a blueprint for financial organisations to harness graph machine learning for actionable insights in transactional ecosystems.
【3】Why and How Auxiliary Tasks Improve JEPA Representations
标题：辅助任务为何以及如何改进JEPA代表
链接：https://arxiv.org/abs/2509.12249
摘要：联合嵌入预测架构（Joint-Embedding Predictive Architecture，JEPA）越来越多地用于视觉表示学习，并作为基于模型的RL的一个组件，但其行为仍然知之甚少。我们提供了一个简单的，实用的JEPA变体，具有辅助回归头共同训练与潜在的动态的理论表征。我们证明了一个无不健康表示崩溃定理：在确定性MDP中，如果训练将潜在转换一致性损失和辅助回归损失都驱动为零，则任何一对非等价观测，即，那些不具有相同的转换动态或辅助标签的特征必须映射到不同的潜在表示。因此，辅助任务锚定表征必须保持的区别。计数环境中的受控消融证实了该理论，并表明与辅助头一起训练JEPA模型比单独训练它们产生更丰富的表示。我们的工作指出了一条改进JEPA编码器的途径：使用辅助函数训练它们，该辅助函数与过渡动态一起编码正确的等价关系。
摘要：Joint-Embedding Predictive Architecture (JEPA) is increasingly used for visual representation learning and as a component in model-based RL, but its behavior remains poorly understood. We provide a theoretical characterization of a simple, practical JEPA variant that has an auxiliary regression head trained jointly with latent dynamics. We prove a No Unhealthy Representation Collapse theorem: in deterministic MDPs, if training drives both the latent-transition consistency loss and the auxiliary regression loss to zero, then any pair of non-equivalent observations, i.e., those that do not have the same transition dynamics or auxiliary label, must map to distinct latent representations. Thus, the auxiliary task anchors which distinctions the representation must preserve. Controlled ablations in a counting environment corroborate the theory and show that training the JEPA model jointly with the auxiliary head generates a richer representation than training them separately. Our work indicates a path to improve JEPA encoders: training them with an auxiliary function that, together with the transition dynamics, encodes the right equivalence relations.
3D|3D重建等相关(2篇)
【1】TripOptimizer: Generative 3D Shape Optimization and Drag Prediction using Triplane VAE Networks
标题：TripOptimizer：使用三平面VAE网络生成式3D形状优化和阻力预测
链接：https://arxiv.org/abs/2509.12224
摘要：传统的基于计算流体动力学的气动外形优化方法计算量大，严重制约了设计空间的拓展。本文介绍了TripOptimizer，这是一个完全可微的深度学习框架，用于直接从车辆点云数据进行快速空气动力学分析和形状优化。TripOptimizer采用变分自动编码器，具有基于三平面的隐式神经表示，用于高保真3D几何重建和阻力系数预测头。在DrivAerNet++上进行训练，DrivAerNet++是一个由8，000个独特车辆几何形状组成的大规模数据集，其相应的阻力系数是通过雷诺平均纳维尔-斯托克斯模拟计算的，该模型学习了一种潜在的表示，对空气动力学显著的几何特征进行编码。我们提出了一种优化策略，修改编码器参数的一个子集，以引导一个初始的几何形状对目标阻力值，并证明其有效性的情况下，优化设计实现阻力系数降低高达11.8%的研究。这些结果随后通过使用超过1.5亿个细胞的独立高保真计算流体动力学模拟进行了验证。隐式表示的一个关键优点是其固有的鲁棒性几何缺陷，使非水密网格的优化，传统的伴随为基础的方法的一个重大挑战。该框架实现了更灵活的气动外形优化工作流程，减少了对计算密集型CFD模拟的依赖，尤其是在早期设计阶段。
摘要：The computational cost of traditional Computational Fluid Dynamics-based Aerodynamic Shape Optimization severely restricts design space exploration. This paper introduces TripOptimizer, a fully differentiable deep learning framework for rapid aerodynamic analysis and shape optimization directly from vehicle point cloud data. TripOptimizer employs a Variational Autoencoder featuring a triplane-based implicit neural representation for high-fidelity 3D geometry reconstruction and a drag coefficient prediction head. Trained on DrivAerNet++, a large-scale dataset of 8,000 unique vehicle geometries with corresponding drag coefficients computed via Reynolds-Averaged Navier-Stokes simulations, the model learns a latent representation that encodes aerodynamically salient geometric features. We propose an optimization strategy that modifies a subset of the encoder parameters to steer an initial geometry towards a target drag value, and demonstrate its efficacy in case studies where optimized designs achieved drag coefficient reductions up to 11.8\%. These results were subsequently validated by using independent, high-fidelity Computational Fluid Dynamics simulations with more than 150 million cells. A key advantage of the implicit representation is its inherent robustness to geometric imperfections, enabling optimization of non-watertight meshes, a significant challenge for traditional adjoint-based methods. The framework enables a more agile Aerodynamic Shape Optimization workflow, reducing reliance on computationally intensive CFD simulations, especially during early design stages.
【2】SamudrACE: Fast and Accurate Coupled Climate Modeling with 3D Ocean and Atmosphere Emulators
标题：SamudrACE：使用3D海洋和大气模拟器快速准确的耦合气候建模
链接：https://arxiv.org/abs/2509.12490
备注：23 pages, 17 figures
摘要：传统的数值全球气候模式通过在大气、海洋、海冰、陆地表面和其他地球物理过程的模拟器之间交换边界条件来模拟整个地球系统。这种范例允许在一个公共框架内分布式开发各个组件，通过耦合器统一起来，该耦合器通过空间或时间对齐和通量交换来处理领域之间的转换。遵循适用于基于机器学习的仿真器的类似方法，我们提出了SamudrACE：一个耦合的全球气候模型仿真器，它可以在1度水平，6小时大气和5天海洋分辨率下进行长达几个世纪的模拟，具有145个二维场，跨越8个大气和19个海洋垂直水平，加上海冰，表面和大气层顶部变量。SamudrACE非常稳定，与具有规定边界强迫的各组成部分相比，气候偏差较低，耦合气候现象（如厄尔尼诺/南方涛动）的实际变异性在非耦合模式下无法模拟。
摘要 ：Traditional numerical global climate models simulate the full Earth system by exchanging boundary conditions between separate simulators of the atmosphere, ocean, sea ice, land surface, and other geophysical processes. This paradigm allows for distributed development of individual components within a common framework, unified by a coupler that handles translation between realms via spatial or temporal alignment and flux exchange. Following a similar approach adapted for machine learning-based emulators, we present SamudrACE: a coupled global climate model emulator which produces centuries-long simulations at 1-degree horizontal, 6-hourly atmospheric, and 5-daily oceanic resolution, with 145 2D fields spanning 8 atmospheric and 19 oceanic vertical levels, plus sea ice, surface, and top-of-atmosphere variables. SamudrACE is highly stable and has low climate biases comparable to those of its components with prescribed boundary forcing, with realistic variability in coupled climate phenomena such as ENSO that is not possible to simulate in uncoupled mode.
编码器(2篇)
【1】Ensemble Visualization With Variational Autoencoder
标题：使用变分自动编码器实现可视化
链接：https://arxiv.org/abs/2509.13000
备注：Accepted by the IEEE Workshop on Uncertainty Visualization
摘要：我们提出了一种新的方法来可视化数据集成，通过构建潜在空间中的结构化概率表示，即，空间数据特征的低维表示。我们的方法转换成一个潜在的空间，通过特征空间转换和无监督学习使用变分自动编码器（VAE）集成的空间特征。由此产生的潜在空间遵循多变量标准高斯分布，从而能够分析计算置信区间和生成数据集合的概率分布的密度估计。天气预报集合的初步结果表明，我们的方法的有效性和通用性。
摘要：We present a new method to visualize data ensembles by constructing structured probabilistic representations in latent spaces, i.e., lower-dimensional representations of spatial data features. Our approach transforms the spatial features of an ensemble into a latent space through feature space conversion and unsupervised learning using a variational autoencoder (VAE). The resulting latent spaces follow multivariate standard Gaussian distributions, enabling analytical computation of confidence intervals and density estimation of the probabilistic distribution that generates the data ensemble. Preliminary results on a weather forecasting ensemble demonstrate the effectiveness and versatility of our method.
【2】VADER: A Variational Autoencoder to Infer Planetary Masses and Gas-Dust Disk Properties Around Young Stars
标题：VADER：一个用于推断年轻恒星周围行星质量和气体尘埃盘属性的变分自动编码器
链接：https://arxiv.org/abs/2509.12324
备注：6 pages, 5 figures, Accepted and Published at International Conference on Machine Learning, Machine Learning for Astrophysics Workshop 2025
摘要：我们提出了\textbf{VADER}（Variational Autoencoder for Embedded with Rings），用于从原行星盘（PPD）的高分辨率ALMA尘埃连续图像推断行星质量和全球盘属性。VADER是一种概率深度学习模型，可以直接从原行星盘图像中对行星质量、$\alpha$-粘度、尘埃与气体之比、斯托克斯数、耀斑指数和行星数量进行不确定性感知推断。VADER在超过100{，}000张PPD合成图像上进行训练，这些图像是通过\texttt{RADMC 3D}后处理的\texttt{FARGO 3D}模拟生成的。我们的训练模型预测物理行星和磁盘参数与$R^2 > 0.9$从尘埃连续图像的PPD。应用到23个真正的磁盘，VADER的质量估计与文献值是一致的，并揭示潜在的相关性，反映已知的磁盘物理。我们的研究结果建立基于VAE的生成模型作为强大的工具，概率天体物理推断，直接应用于解释原行星盘子结构的时代，大型干涉测量。
摘要：We present \textbf{VADER} (Variational Autoencoder for Disks Embedded with Rings), for inferring both planet mass and global disk properties from high-resolution ALMA dust continuum images of protoplanetary disks (PPDs). VADER, a probabilistic deep learning model, enables uncertainty-aware inference of planet masses, $\alpha$-viscosity, dust-to-gas ratio, Stokes number, flaring index, and the number of planets directly from protoplanetary disk images. VADER is trained on over 100{,}000 synthetic images of PPDs generated from \texttt{FARGO3D} simulations post-processed with \texttt{RADMC3D}. Our trained model predicts physical planet and disk parameters with $R^2 > 0.9$ from dust continuum images of PPDs. Applied to 23 real disks, VADER's mass estimates are consistent with literature values and reveal latent correlations that reflect known disk physics. Our results establish VAE-based generative models as robust tools for probabilistic astrophysical inference, with direct applications to interpreting protoplanetary disk substructures in the era of large interferometric surveys.
优化|敛散性(4篇)
【1】Single-stream Policy Optimization
标题：单流政策优化
链接：https://arxiv.org/abs/2509.13232
摘要：我们从单流的角度重新审视大型语言模型（LLM）的策略梯度优化。流行的基于组的方法，如GRPO，减少了与动态基线的差异，但有严重的缺陷：频繁的退化组会消除学习信号，同步障碍会阻碍可扩展性。我们引入了单流策略优化（SPO），它通过设计消除了这些问题。SPO用一个持久的、KL自适应的值跟踪器取代了每个组的基线，并在整个批次中对优势进行了全局标准化，为每个样本提供了稳定的、低方差的学习信号。由于没有分组，SPO可以实现更高的吞吐量，并在生成时间不同的长期或工具集成设置中有效地扩展。此外，持久的价值跟踪器自然能够通过优先级采样实现自适应课程。在Qwen 3 -8B上的实验表明，SPO算法比GRPO算法收敛更平滑，精度更高，同时消除了退化群上的计算浪费.消融研究证实，SPO的收益源于其基线估计和优势归一化的原则性方法，为LLM推理提供了一条更强大、更有效的路径。在Qwen 3 8B的五个硬数学基准测试中，SPO将平均maj@32比GRPO提高了+3.4个百分点（pp），这是由具有挑战性的数据集上的大量绝对点增益驱动的，包括BRUMO 25上的+7.3 pp，AIME 25上的+4.4 pp，HMMT 25上的+3.3 pp，并在评估的$k$值中实现了一致的相对增益。SPO的成功挑战了为RL算法增加偶然复杂性的流行趋势，突出了一条基本原则而不是架构解决方案推动LLM推理下一波进步的道路。
摘要：We revisit policy-gradient optimization for Large Language Models (LLMs) from a single-stream perspective. Prevailing group-based methods like GRPO reduce variance with on-the-fly baselines but suffer from critical flaws: frequent degenerate groups erase learning signals, and synchronization barriers hinder scalability. We introduce Single-stream Policy Optimization (SPO), which eliminates these issues by design. SPO replaces per-group baselines with a persistent, KL-adaptive value tracker and normalizes advantages globally across the batch, providing a stable, low-variance learning signal for every sample. Being group-free, SPO enables higher throughput and scales effectively in long-horizon or tool-integrated settings where generation times vary. Furthermore, the persistent value tracker naturally enables an adaptive curriculum via prioritized sampling. Experiments using Qwen3-8B show that SPO converges more smoothly and attains higher accuracy than GRPO, while eliminating computation wasted on degenerate groups. Ablation studies confirm that SPO's gains stem from its principled approach to baseline estimation and advantage normalization, offering a more robust and efficient path for LLM reasoning. Across five hard math benchmarks with Qwen3 8B, SPO improves the average maj@32 by +3.4 percentage points (pp) over GRPO, driven by substantial absolute point gains on challenging datasets, including +7.3 pp on BRUMO 25, +4.4 pp on AIME 25, +3.3 pp on HMMT 25, and achieves consistent relative gain in pass@$k$ across the evaluated $k$ values. SPO's success challenges the prevailing trend of adding incidental complexity to RL algorithms, highlighting a path where fundamental principles, not architectural workarounds, drive the next wave of progress in LLM reasoning.
【2】EmbeddedML: A New Optimized and Fast Machine Learning Library
标题：EmbeddedML：一个新的优化且快速的机器学习库
链接：https://arxiv.org/abs/2509.12774
备注：10 pages, 7 figures
摘要：机器学习模型和库可以训练不同大小的数据集，并执行预测和分类操作，但机器学习模型和库会导致大型数据集的训练时间缓慢而长。本文介绍了EmbeddedML，一个训练时间优化和数学增强的机器学习库。与scikit-learn相比，速度提高了大约一倍，而在回归模型（如多元线性回归）的准确性方面没有任何损失。逻辑回归和支持向量机（SVM）算法已经在数学上重写，以减少训练时间并提高分类模型的准确性。通过应用数学改进，与scikit-learn实现相比，SVM在小数据集上的训练时间减少了约2倍，在大数据集上减少了约800倍，Logistic回归减少了约4倍。总之，EmbeddedML库提供了回归、分类、聚类和降维算法，这些算法经过数学重写和优化以减少训练时间。
摘要：Machine learning models and libraries can train datasets of different sizes and perform prediction and classification operations, but machine learning models and libraries cause slow and long training times on large datasets. This article introduces EmbeddedML, a training-time-optimized and mathematically enhanced machine learning library. The speed was increased by approximately times compared to scikit-learn without any loss in terms of accuracy in regression models such as Multiple Linear Regression. Logistic Regression and Support Vector Machines (SVM) algorithms have been mathematically rewritten to reduce training time and increase accuracy in classification models. With the applied mathematical improvements, training time has been reduced by approximately 2 times for SVM on small datasets and by around 800 times on large datasets, and by approximately 4 times for Logistic Regression, compared to the scikit-learn implementation. In summary, the EmbeddedML library offers regression, classification, clustering, and dimensionality reduction algorithms that are mathematically rewritten and optimized to reduce training time.
【3】Integrating Attention-Enhanced LSTM and Particle Swarm Optimization for Dynamic Pricing and Replenishment Strategies in Fresh Food Supermarkets
标题：集成注意力增强LSTM和粒子群优化，用于生鲜食品超市的动态定价和补充策略
链接：https://arxiv.org/abs/2509.12339
备注：16 pages, 6 figure
摘要：本文提出了一种新的方法，通过结合长短期记忆（LSTM）网络和粒子群优化（PSO）来优化生鲜超市的定价和补货策略。LSTM模型通过注意力机制增强，用于预测七天内的销售量、定价趋势和变质率。LSTM模型生成的预测作为PSO算法的输入，PSO算法迭代优化定价和补货策略，以在遵守库存约束的同时最大化盈利能力。成本加成定价的整合允许根据固定和可变成本进行动态调整，确保对市场波动的实时适应性。该框架不仅能最大限度地提高利润，还能减少食物浪费，为超市的可持续运营做出贡献。注意力机制通过识别关键时间点和影响销售的因素来增强LSTM模型的可解释性，提高决策准确性。这种方法弥合了预测建模和优化之间的差距，为新鲜食品零售和其他易腐商品行业的动态定价和库存管理提供了可扩展的解决方案。
摘要：This paper presents a novel approach to optimizing pricing and replenishment strategies in fresh food supermarkets by combining Long Short-Term Memory (LSTM) networks with Particle Swarm Optimization (PSO). The LSTM model, enhanced with an attention mechanism, is used to predict sales volumes, pricing trends, and spoilage rates over a seven-day period. The predictions generated by the LSTM model serve as inputs for the PSO algorithm, which iteratively optimizes pricing and replenishment strategies to maximize profitability while adhering to inventory constraints. The integration of cost-plus pricing allows for dynamic adjustments based on fixed and variable costs, ensuring real-time adaptability to market fluctuations. The framework not only maximizes profits but also reduces food waste, contributing to more sustainable supermarket operations. The attention mechanism enhances the interpretability of the LSTM model by identifying key time points and factors influencing sales, improving decision-making accuracy. This methodology bridges the gap between predictive modeling and optimization, offering a scalable solution for dynamic pricing and inventory management in fresh food retail and other industries dealing with perishable goods.
【4】DeltaHedge: A Multi-Agent Framework for Portfolio Options Optimization
标题：Delta Hedge：投资组合期权优化的多代理框架
链接：https://arxiv.org/abs/2509.12753
备注：Presented at Pacific Asia Conference on Information Systems (PACIS 2025), Kuala Lumpur. Official proceedings available at https://aisel.aisnet.org/pacis2025/aiandml/aiandml/25/. 16 pages, 7 figures, 3 tables
摘要：在动荡的金融市场中，平衡风险和回报仍然是一项重大挑战。传统的方法往往只关注股权配置，忽视了期权交易在动态风险对冲方面的战略优势。这项工作提出了DeltaHedge，这是一个多代理框架，将期权交易与AI驱动的投资组合管理集成在一起。通过将先进的强化学习技术与基于期权的对冲策略相结合，DeltaHedge提高了风险调整后的回报，并在不同的市场条件下稳定了投资组合的表现。实验结果表明，DeltaHedge优于传统的策略和独立的模型，强调其潜力，以改变实际的投资组合管理在复杂的金融环境。基于这些发现，本文通过引入一种用于整合期权交易策略的新型多代理系统，为量化金融和人工智能驱动的投资组合优化领域做出了贡献，解决了现有文献中的一个空白。
摘要：In volatile financial markets, balancing risk and return remains a significant challenge. Traditional approaches often focus solely on equity allocation, overlooking the strategic advantages of options trading for dynamic risk hedging. This work presents DeltaHedge, a multi-agent framework that integrates options trading with AI-driven portfolio management. By combining advanced reinforcement learning techniques with an ensembled options-based hedging strategy, DeltaHedge enhances risk-adjusted returns and stabilizes portfolio performance across varying market conditions. Experimental results demonstrate that DeltaHedge outperforms traditional strategies and standalone models, underscoring its potential to transform practical portfolio management in complex financial environments. Building on these findings, this paper contributes to the fields of quantitative finance and AI-driven portfolio optimization by introducing a novel multi-agent system for integrating options trading strategies, addressing a gap in the existing literature.
预测|估计(10篇)
【1】On the Correlation between Individual Fairness and Predictive Accuracy in Probabilistic Models
标题：概率模型中个体公平性与预测准确性之间的相关性
链接：https://arxiv.org/abs/2509.13165
备注：15 pages, 9 figures, 1 table
摘要：我们调查个人公平性生成概率分类器通过分析的鲁棒性后验推理扰动私人功能。在稳健性分析的基础上，我们假设稳健性和预测准确性之间存在相关性，具体而言，表现出更高稳健性的实例更有可能被准确分类。我们经验性地评估这一假设，使用基准的14个数据集的公平性问题，采用贝叶斯网络作为底层的生成模型。为了解决与多个私有特征的贝叶斯网络的鲁棒性分析的计算复杂性，我们重新制定的问题作为一个最可能的解释任务，在辅助马尔可夫随机场。我们的实验证实了相关性的假设，提出了新的方向，以减轻传统的公平性和准确性之间的权衡。
摘要：We investigate individual fairness in generative probabilistic classifiers by analysing the robustness of posterior inferences to perturbations in private features. Building on established results in robustness analysis, we hypothesise a correlation between robustness and predictive accuracy, specifically, instances exhibiting greater robustness are more likely to be classified accurately. We empirically assess this hypothesis using a benchmark of fourteen datasets with fairness concerns, employing Bayesian networks as the underlying generative models. To address the computational complexity associated with robustness analysis over multiple private features with Bayesian networks, we reformulate the problem as a most probable explanation task in an auxiliary Markov random field. Our experiments confirm the hypothesis about the correlation, suggesting novel directions to mitigate the traditional trade-off between fairness and accuracy.
【2】Deep Learning for Model-Free Prediction of Thermal States of Robot Joint Motors
标题：深度学习用于机器人关节电机热状态的无模型预测
链接：https://arxiv.org/abs/2509.12739
备注：$©$ 2025 the authors. This work has been accepted to the 10th IFAC Symposium on Mechatronic Systems & 14th IFAC Symposium on Robotics July 15-18, 2025 || Paris, France for publication under a Creative Commons Licence CC-BY-NC-ND
摘要：在这项工作中，由多个隐藏的长短期记忆（LSTM）和前馈层组成的深度神经网络被训练来预测机器人操纵器关节电机的热行为。采用了无模型和可扩展的方法。它适应复杂性和不确定性的挑战，源于推导，识别和验证的大量参数的近似模型，这是很难获得的。为此，收集并处理感测到的关节扭矩以预见关节马达的热行为。提出了基于机器学习的七关节冗余度机器人关节电机温度动态捕捉的预测结果。
摘要：In this work, deep neural networks made up of multiple hidden Long Short-Term Memory (LSTM) and Feedforward layers are trained to predict the thermal behavior of the joint motors of robot manipulators. A model-free and scalable approach is adopted. It accommodates complexity and uncertainty challenges stemming from the derivation, identification, and validation of a large number of parameters of an approximation model that is hardly available. To this end, sensed joint torques are collected and processed to foresee the thermal behavior of joint motors. Promising prediction results of the machine learning based capture of the temperature dynamics of joint motors of a redundant robot with seven joints are presented.
【3】A Novel Recurrent Neural Network Framework for Prediction and Treatment of Oncogenic Mutation Progression
标题：用于预测和治疗致癌突变进展的新型回归神经网络框架
链接：https://arxiv.org/abs/2509.12732
备注：12 pages, 11 figures, work originally done in 2022/2023 and was awarded as one of the Regeneron Science Talent Search Finalists in 2022
摘要：尽管医学取得了重大进步，但癌症仍然是第二大死亡原因，在美国每年有超过60万人死亡。一个新兴的领域，途径分析，是有前途的，但仍然依赖于手动获得的湿实验室数据，这是耗时的获取。这项工作为基于人工智能（AI）的通路分析提出了一个高效、有效的端到端框架，可以预测癌症的严重程度和突变进展，从而推荐可能的治疗方法。所提出的技术涉及时间序列机器学习模型和路径分析的新组合。首先，从癌症基因组图谱（TCGA）数据库中分离突变序列。然后，采用一种新的预处理算法，根据变异频率过滤关键变异。这些数据被输入到预测癌症严重程度的递归神经网络（RNN）中。然后，该模型概率性地使用RNN预测、来自预处理算法的信息和多个药物靶标数据库来预测未来的突变并推荐可能的治疗方法。该框架实现了稳健的结果和受试者操作特征（ROC）曲线（一个关键的统计指标），准确率超过60%，类似于现有的癌症诊断。此外，预处理在分离重要突变方面发挥了重要作用，表明研究的每个癌症阶段可能包含数百个关键驱动突变，与当前研究一致。还生成了基于预测基因频率的热图，突出显示了每种癌症中的关键突变。总的来说，这项工作是第一个提出一个有效的，具有成本效益的端到端框架，用于预测癌症进展并提供可能的治疗方法，而不依赖于昂贵，耗时的湿实验室工作。
摘要：Despite significant medical advancements, cancer remains the second leading cause of death, with over 600,000 deaths per year in the US. One emerging field, pathway analysis, is promising but still relies on manually derived wet lab data, which is time-consuming to acquire. This work proposes an efficient, effective end-to-end framework for Artificial Intelligence (AI) based pathway analysis that predicts both cancer severity and mutation progression, thus recommending possible treatments. The proposed technique involves a novel combination of time-series machine learning models and pathway analysis. First, mutation sequences were isolated from The Cancer Genome Atlas (TCGA) Database. Then, a novel preprocessing algorithm was used to filter key mutations by mutation frequency. This data was fed into a Recurrent Neural Network (RNN) that predicted cancer severity. Then, the model probabilistically used the RNN predictions, information from the preprocessing algorithm, and multiple drug-target databases to predict future mutations and recommend possible treatments. This framework achieved robust results and Receiver Operating Characteristic (ROC) curves (a key statistical metric) with accuracies greater than 60%, similar to existing cancer diagnostics. In addition, preprocessing played an instrumental role in isolating important mutations, demonstrating that each cancer stage studied may contain on the order of a few-hundred key driver mutations, consistent with current research. Heatmaps based on predicted gene frequency were also generated, highlighting key mutations in each cancer. Overall, this work is the first to propose an efficient, cost-effective end-to-end framework for projecting cancer progression and providing possible treatments without relying on expensive, time-consuming wet lab work.
【4】Spatio-temporal DeepKriging in PyTorch: A Supplementary Application to Precipitation Data for Interpolation and Probabilistic Forecasting
标题：PyTorch中的时空DeepKriging：降水数据用于内插和概率预测的补充应用
链接：https://arxiv.org/abs/2509.12708
摘要：详细分析了欧洲的降水数据，重点是插值和预测应用。时空DeepKriging（STDK）框架已使用PyTorch平台实现，以实现这些目标。该模型能够处理时空不规则性，同时生成高分辨率插值和多步预测。可复制的代码模块已经被开发为独立的PyTorch实现，用于插值\footnote[2]{Interpolation -https：//github.com/pratiknag/Spatio-temporalDeepKriging-Pytorch.git}和预测\footnote[3]{Forecasting -https：//github.com/pratiknag/pytorch-convlstm.git}，促进了类似气候数据集的更广泛应用。该方法的有效性通过对每日降水量测量的广泛评估来证明，突出了预测性能和鲁棒性。
摘要：A detailed analysis of precipitation data over Europe is presented, with a focus on interpolation and forecasting applications. A Spatio-temporal DeepKriging (STDK) framework has been implemented using the PyTorch platform to achieve these objectives. The proposed model is capable of handling spatio-temporal irregularities while generating high-resolution interpolations and multi-step forecasts. Reproducible code modules have been developed as standalone PyTorch implementations for the interpolation\footnote[2]{Interpolation - https://github.com/pratiknag/Spatio-temporalDeepKriging-Pytorch.git} and forecasting\footnote[3]{Forecasting - https://github.com/pratiknag/pytorch-convlstm.git}, facilitating broader application to similar climate datasets. The effectiveness of this approach is demonstrated through extensive evaluation on daily precipitation measurements, highlighting predictive performance and robustness.
【5】Mob-based cattle weight gain forecasting using ML models
标题：使用ML模型预测基于Mo的牛增重
链接：https://arxiv.org/abs/2509.12615
备注：None
摘要：基于群体的牛增重预测（MB CWG）可能会使大型畜牧场受益，使农民能够改进其饲养策略，做出明智的育种选择，并降低与气候变化和市场波动相关的风险。在本文中，提出了一种新的技术，称为MB CWG预测一个月的先进的体重增加的牛群为基础的牛使用历史数据收集的查尔斯特大学农场。本研究采用随机森林（RF）模型，比较其与支持向量回归（SVR）和长短期记忆（LSTM）模型在每月体重增长预测方面的性能。四个数据集被用来评估模型的性能，使用756个样本数据，从108头牛，随着天气数据（降雨量和温度）影响CWG。RF模型在所有数据集上的表现都优于SVR和LSTM模型，当包括天气和年龄因素时，R^2为0.973，RMSE为0.040，MAE为0.033。结果表明，包括天气和年龄因素显着提高了体重增加预测的准确性，RF模型在所有情况下都优于SVR和LSTM模型。这些研究结果表明，RF作为一种强大的工具，在可变条件下预测牛的体重增长的潜力，突出了年龄和气候因素对牛群体重趋势的影响。本研究还开发了一种创新的自动化预处理工具，为MB CWG预测模型生成基准数据集。该工具在GitHub上公开提供，可以帮助为当前和未来的分析研究准备数据集。
摘要：Forecasting mob based cattle weight gain (MB CWG) may benefit large livestock farms, allowing farmers to refine their feeding strategies, make educated breeding choices, and reduce risks linked to climate variability and market fluctuations. In this paper, a novel technique termed MB CWG is proposed to forecast the one month advanced weight gain of herd based cattle using historical data collected from the Charles Sturt University Farm. This research employs a Random Forest (RF) model, comparing its performance against Support Vector Regression (SVR) and Long Short Term Memory (LSTM) models for monthly weight gain prediction. Four datasets were used to evaluate the performance of models, using 756 sample data from 108 herd-based cattle, along with weather data (rainfall and temperature) influencing CWG. The RF model performs better than the SVR and LSTM models across all datasets, achieving an R^2 of 0.973, RMSE of 0.040, and MAE of 0.033 when both weather and age factors were included. The results indicate that including both weather and age factors significantly improves the accuracy of weight gain predictions, with the RF model outperforming the SVR and LSTM models in all scenarios. These findings demonstrate the potential of RF as a robust tool for forecasting cattle weight gain in variable conditions, highlighting the influence of age and climatic factors on herd based weight trends. This study has also developed an innovative automated pre processing tool to generate a benchmark dataset for MB CWG predictive models. The tool is publicly available on GitHub and can assist in preparing datasets for current and future analytical research..
【6】No Need for "Learning" to Defer? A Training Free Deferral Framework to Multiple Experts through Conformal Prediction
链接：https://arxiv.org/abs/2509.12573
备注：9 pages, 4 figures, 1 table
摘要：人工智能系统通常无法在所有输入中提供可靠的预测，这促使人们需要混合人工智能决策。现有的学习延迟（L2 D）方法通过训练延迟模型来解决这个问题，但是这些方法对专家组成的变化很敏感，如果专家发生变化，则需要进行大量的再培训。我们提出了一个无训练，模型和专家不可知的框架，专家推迟共形预测的基础上。我们的方法使用由共形预测器生成的预测集来识别特定于标签的不确定性，并使用隔离标准选择最具鉴别力的专家，测量专家如何区分剩余的合理标签。在CIFAR 10-H和ImageNet 16-H上的实验表明，我们的方法始终优于独立模型和最强专家，准确率分别达到99.57\pm0.10\%$和99.40\pm0.52\%$，同时将专家工作量减少了11 $。该方法在专家性能下降的情况下仍然具有鲁棒性，并且在低信息设置中表现出逐渐的性能下降。这些结果为现实世界的人类-AI协作提供了一种可扩展的、无需再训练的L2 D替代方案。
摘要：AI systems often fail to deliver reliable predictions across all inputs, prompting the need for hybrid human-AI decision-making. Existing Learning to Defer (L2D) approaches address this by training deferral models, but these are sensitive to changes in expert composition and require significant retraining if experts change. We propose a training-free, model- and expert-agnostic framework for expert deferral based on conformal prediction. Our method uses the prediction set generated by a conformal predictor to identify label-specific uncertainty and selects the most discriminative expert using a segregativity criterion, measuring how well an expert distinguishes between the remaining plausible labels. Experiments on CIFAR10-H and ImageNet16-H show that our method consistently outperforms both the standalone model and the strongest expert, with accuracies attaining $99.57\pm0.10\%$ and $99.40\pm0.52\%$, while reducing expert workload by up to a factor of $11$. The method remains robust under degraded expert performance and shows a gradual performance drop in low-information settings. These results suggest a scalable, retraining-free alternative to L2D for real-world human-AI collaboration.
【7】C3DE: Causal-Aware Collaborative Neural Controlled Differential Equation for Long-Term Urban Crowd Flow Prediction
标题：C3 DE：用于长期城市人群流量预测的Cause-Aware协作神经控制方程
链接：https://arxiv.org/abs/2509.12289
摘要：由于序列长度和采样间隔的增加，长期的城市人群流量预测会受到累积采样误差的严重影响，这激发了我们利用神经控制微分方程（NCDE）来缓解这个问题。然而，由于兴趣点（POI）的演化对长期人群流的影响至关重要，人群流和POI分布之间的多时间尺度异步动态关系以及潜在的虚假因果关系，使得NCDE应用于长期城市人群流预测面临挑战。为此，我们提出了Cause-aware协作神经CDE（C3 DE）来模拟人群流的长期动态。具体来说，我们引入了一个双路径NCDE作为骨干，以有效地捕捉跨多个时间尺度的协作信号的异步演化。然后，我们设计了一个动态的校正机制与反事实的因果效应估计量化的因果影响的兴趣点对人群流量，并尽量减少虚假相关性的积累。最后，我们利用预测器进行长期预测，融合POI和人群流量的协同信号。在三个真实世界数据集上进行的大量实验表明，C3 DE的性能优越，特别是在流量波动明显的城市。
摘要：Long-term urban crowd flow prediction suffers significantly from cumulative sampling errors, due to increased sequence lengths and sampling intervals, which inspired us to leverage Neural Controlled Differential Equations (NCDEs) to mitigate this issue. However, regarding the crucial influence of Points of Interest (POIs) evolution on long-term crowd flow, the multi-timescale asynchronous dynamics between crowd flow and POI distribution, coupled with latent spurious causality, poses challenges to applying NCDEs for long-term urban crowd flow prediction. To this end, we propose Causal-aware Collaborative neural CDE (C3DE) to model the long-term dynamic of crowd flow. Specifically, we introduce a dual-path NCDE as the backbone to effectively capture the asynchronous evolution of collaborative signals across multiple time scales. Then, we design a dynamic correction mechanism with the counterfactual-based causal effect estimator to quantify the causal impact of POIs on crowd flow and minimize the accumulation of spurious correlations. Finally, we leverage a predictor for long-term prediction with the fused collaborative signals of POI and crowd flow. Extensive experiments on three real-world datasets demonstrate the superior performance of C3DE, particularly in cities with notable flow fluctuations.
【8】Deriving the Scaled-Dot-Function via Maximum Likelihood Estimation and Maximum Entropy Approach
标题：通过最大似然估计和最大熵方法推导标度点函数
链接：https://arxiv.org/abs/2509.12285
摘要：在本文中，我们提出了一个最大似然估计方法来确定值向量的Transformer模型。我们将值向量、键向量和查询向量的序列建模为高斯分布序列。每个高斯分布中的方差取决于时间步长、对应的键向量和查询向量。每个高斯分布中的平均值取决于时间步长和相应的值向量。该分析可以为Transformer架构中使用的缩放点积函数或softmax函数提供新的解释[1]。受[4]启发的另一种解释是基于自然语言处理中的最大熵方法[5]。在这种方法中，查询向量和关键向量被用来获得最大熵模型的特征函数。
摘要 ：In this paper, we present a maximum likelihood estimation approach to determine the value vector in transformer models. We model the sequence of value vectors, key vectors, and the query vector as a sequence of Gaussian distributions. The variance in each Gaussian distribution depends on the time step, the corresponding key vector, and the query vector. The mean value in each Gaussian distribution depends on the time step, and the corresponding value vector. This analysis may offer a new explanation of the scaled-dot-product function or softmax function used in transformer architectures [1]. Another explanation, inspired by [4], is based on the maximum entropy approach in natural language processing [5]. In this approach, a query vector and key vectors are used to derive the feature functions for the maximum entropy model.
【9】Neural Diffeomorphic-Neural Operator for Residual Stress-Induced Deformation Prediction
标题：残余应力致变形预测的神经同形神经操作器
链接：https://arxiv.org/abs/2509.12237
摘要：准确预测结构件的加工变形是保证结构件尺寸精度和可靠性的关键。这种变形通常源于残余应力场，其分布和影响随几何复杂性而显著变化。传统的模拟残余应力和变形之间的耦合的数值方法是计算昂贵的，特别是当考虑不同的几何形状。神经操作符最近出现作为一个强大的范例，有效地解决偏微分方程，提供显着的优势，在加速残余应力-变形分析。然而，它们在不断变化的几何域中的直接应用面临着理论和实践的限制。为了解决这个问题，一个新的框架，基于同构嵌入神经算子命名为神经同构神经算子（NDNO）。复杂的三维几何图形显式映射到一个共同的参考域通过一个同构的神经网络约束的光滑性和可逆性。然后，在这个参考域上训练神经运算符，从而能够有效地学习由残余应力引起的变形场。一旦经过训练，同构神经网络和神经运算符都表现出高效的预测能力，可以快速适应不同的几何形状。因此，所提出的方法提供了一个有效的和计算效率高的解决方案，变形预测的结构部件受到不同的几何形状。所提出的方法被验证，以预测主方向和多方向的变形场，实现高精度和高效率的零件具有不同的几何形状，包括组件类型，尺寸和功能。
摘要：Accurate prediction of machining deformation in structural components is essential for ensuring dimensional precision and reliability. Such deformation often originates from residual stress fields, whose distribution and influence vary significantly with geometric complexity. Conventional numerical methods for modeling the coupling between residual stresses and deformation are computationally expensive, particularly when diverse geometries are considered. Neural operators have recently emerged as a powerful paradigm for efficiently solving partial differential equations, offering notable advantages in accelerating residual stress-deformation analysis. However, their direct application across changing geometric domains faces theoretical and practical limitations. To address this challenge, a novel framework based on diffeomorphic embedding neural operators named neural diffeomorphic-neural operator (NDNO) is introduced. Complex three-dimensional geometries are explicitly mapped to a common reference domain through a diffeomorphic neural network constrained by smoothness and invertibility. The neural operator is then trained on this reference domain, enabling efficient learning of deformation fields induced by residual stresses. Once trained, both the diffeomorphic neural network and the neural operator demonstrate efficient prediction capabilities, allowing rapid adaptation to varying geometries. The proposed method thus provides an effective and computationally efficient solution for deformation prediction in structural components subject to varying geometries. The proposed method is validated to predict both main-direction and multi-direction deformation fields, achieving high accuracy and efficiency across parts with diverse geometries including component types, dimensions and features.
【10】A Physics-Informed Neural Networks-Based Model Predictive Control Framework for $SIR$ Epidemics
标题：针对$Sir$流行病的基于物理信息的神经网络的模型预测控制框架
链接：https://arxiv.org/abs/2509.12226
摘要：本文介绍了一种基于物理信息神经网络的模型预测控制框架，用于描述可感染-恢复（$SIR$）传播模型。在流行病控制的MPC设计中的现有研究通常假设1）动态的可测量状态，其中参数被学习，或者2）模型的已知参数，其中状态被学习。在这项工作中，我们解决的联合实时估计的状态和参数的MPC框架内，仅使用噪声感染状态，假设1）只有恢复率是已知的，或2）只有基本的再生数是已知的。在第一个假设下，我们提出了MPC-PINNs和两个新的PINNs算法，所有这些都集成到MPC框架。首先，我们介绍MPC PINN，这是专为$SIR$模型与控制。然后，我们提出了对数缩放PINN（MPC-LS-PINN），它包含一个对数缩放的损失函数，以提高对噪声的鲁棒性。接下来，我们提出了分裂积分PINNs（MPC-SI-PINNs），它利用神经网络训练过程中的积分算子和状态耦合来有效地重建完整的流行病状态信息。在这些方法的基础上，我们进一步扩展了第二个假设的框架。我们建立了必要的条件，并扩展了我们的PINNs算法，其中MPC-SI-PINNs简化为分裂PINNs（MPC-S-PINNs）。通过将这些算法的MPC框架，我们同时估计流行病的状态和参数，同时产生最优的控制策略。实验结果证明了所提出的方法在不同环境下的有效性。
摘要：This work introduces a physics-informed neural networks (PINNs)-based model predictive control (MPC) framework for susceptible-infected-recovered ($SIR$) spreading models. Existing studies in MPC design for epidemic control often assume either 1) measurable states of the dynamics, where the parameters are learned, or 2) known parameters of the model, where the states are learned. In this work, we address the joint real-time estimation of states and parameters within the MPC framework using only noisy infected states, under the assumption that 1) only the recovery rate is known, or 2) only the basic reproduction number is known. Under the first assumption, we propose MPC-PINNs and two novel PINNs algorithms, all of which are integrated into the MPC framework. First, we introduce MPC-PINNs, which are designed for $SIR$ models with control. We then propose log-scaled PINNs (MPC-LS-PINNs), which incorporate a log-scaled loss function to improve robustness against noise. Next, we present split-integral PINNs (MPC-SI-PINNs), which leverage integral operators and state coupling in the neural network training process to effectively reconstruct the complete epidemic state information. Building upon these methods, we further extend our framework for the second assumption. We establish the necessary conditions and extend our PINNs algorithms, where MPC-SI-PINNs are simplified as split-PINNs (MPC-S-PINNs). By incorporating these algorithms into the MPC framework, we simultaneously estimate the epidemic states and parameters while generating optimal control strategies. Experiment results demonstrate the effectiveness of the proposed methods under different settings.
其他神经网络|深度学习|模型|建模(17篇)
【1】FOSSIL: Regret-minimizing weighting for robust learning under imbalance and small data
标题：FOSSIL：在不平衡和小数据下实现稳健学习的遗憾最小化加权
链接：https://arxiv.org/abs/2509.13218
备注：24 pages, 6 figures, submitted to ICLR 2025
摘要：不平衡和小数据机制在罕见疾病成像、基因组学和灾难响应等领域中普遍存在，其中标记样本稀缺，并且朴素增强通常会引入伪影。现有的解决方案，如过采样、焦点丢失或元加权，解决了这一挑战的孤立方面，但仍然脆弱或复杂。我们引入了FOSSIL（通过样本敏感重要性学习进行灵活优化），这是一个统一的加权框架，可以将类不平衡校正，难度感知课程，增强惩罚和热身动态无缝集成到一个可解释的公式中。与先前的算法不同，所提出的框架提供了基于遗憾的理论保证，并在合成和真实世界数据集上实现了与ERM，课程和元权重基线一致的经验收益，同时不需要架构更改。
摘要 ：Imbalanced and small data regimes are pervasive in domains such as rare disease imaging, genomics, and disaster response, where labeled samples are scarce and naive augmentation often introduces artifacts. Existing solutions such as oversampling, focal loss, or meta-weighting address isolated aspects of this challenge but remain fragile or complex. We introduce FOSSIL (Flexible Optimization via Sample Sensitive Importance Learning), a unified weighting framework that seamlessly integrates class imbalance correction, difficulty-aware curricula, augmentation penalties, and warmup dynamics into a single interpretable formula. Unlike prior heuristics, the proposed framework provides regret-based theoretical guarantees and achieves consistent empirical gains over ERM, curriculum, and meta-weighting baselines on synthetic and real-world datasets, while requiring no architectural changes.
【2】CoVariance Filters and Neural Networks over Hilbert Spaces
标题：Hilbert空间上的协方差过滤器和神经网络
链接：https://arxiv.org/abs/2509.13178
备注：6 pages, 3 figures
摘要：协方差神经网络（VNN）在有限维希尔伯特空间上定义的信号的经验协方差矩阵上执行图卷积，其动机是鲁棒性和可转移性。然而，很少有人知道这些论点如何扩展到无限维希尔伯特空间。在这项工作中，我们迈出了第一步，为定义在无限维希尔伯特空间上的信号引入了一种新的卷积学习框架，以（经验）协方差算子为中心。我们建设性地定义了希尔伯特协方差滤波器（HVFs）和设计希尔伯特协方差网络（HVNs）作为堆栈的HVF滤波器组与非线性激活。我们提出了一个原则性的离散化过程，我们证明了经验HVFs可以恢复滤波信号的功能PCA（FPCA）。然后，我们描述了我们的框架的多功能性，从多变量实值函数再生核希尔伯特空间的例子。最后，我们在合成和真实世界的时间序列分类任务上验证了HVNs，与基于MLP和FPCA的分类器相比，表现出强大的性能。
摘要：CoVariance Neural Networks (VNNs) perform graph convolutions on the empirical covariance matrix of signals defined over finite-dimensional Hilbert spaces, motivated by robustness and transferability properties. Yet, little is known about how these arguments extend to infinite-dimensional Hilbert spaces. In this work, we take a first step by introducing a novel convolutional learning framework for signals defined over infinite-dimensional Hilbert spaces, centered on the (empirical) covariance operator. We constructively define Hilbert coVariance Filters (HVFs) and design Hilbert coVariance Networks (HVNs) as stacks of HVF filterbanks with nonlinear activations. We propose a principled discretization procedure, and we prove that empirical HVFs can recover the Functional PCA (FPCA) of the filtered signals. We then describe the versatility of our framework with examples ranging from multivariate real-valued functions to reproducing kernel Hilbert spaces. Finally, we validate HVNs on both synthetic and real-world time-series classification tasks, showing robust performance compared to MLP and FPCA-based classifiers.
【3】Curriculum Learning for Mesh-based simulations
标题：基于网格的模拟的课程学习
链接：https://arxiv.org/abs/2509.13138
摘要：图神经网络（GNN）已经成为基于网格的计算流体动力学（CFD）的强大替代品，但在具有数十万节点的高分辨率非结构化网格上训练它们仍然非常昂贵。我们研究了一个由粗到细的课程，它通过首先在非常粗糙的网格上训练，然后逐步引入中等和高分辨率（最多3\times10^5\）节点来加速收敛。与多尺度GNN架构不同，模型本身是不变的;只有训练数据的保真度随时间变化。我们实现了相当的泛化精度，同时减少了高达50%的总挂钟时间。此外，在我们的模型缺乏学习基础物理的能力的数据集上，使用课程学习使其能够突破高原。
摘要：Graph neural networks (GNNs) have emerged as powerful surrogates for mesh-based computational fluid dynamics (CFD), but training them on high-resolution unstructured meshes with hundreds of thousands of nodes remains prohibitively expensive. We study a \emph{coarse-to-fine curriculum} that accelerates convergence by first training on very coarse meshes and then progressively introducing medium and high resolutions (up to \(3\times10^5\) nodes). Unlike multiscale GNN architectures, the model itself is unchanged; only the fidelity of the training data varies over time. We achieve comparable generalization accuracy while reducing total wall-clock time by up to 50\%. Furthermore, on datasets where our model lacks the capacity to learn the underlying physics, using curriculum learning enables it to break through plateaus.
【4】Sublinear-Time Algorithms for Diagonally Dominant Systems and Applications to the Friedkin-Johnsen Model
标题：对角优势系统的次线性时间算法及其在Friedkin-Johnsen模型中的应用
链接：https://arxiv.org/abs/2509.13112
摘要：We study sublinear-time algorithms for solving linear systems $Sz = b$, where $S$ is a diagonally dominant matrix, i.e., $|S_{ii}| \geq \delta + \sum_{j \ne i} |S_{ij}|$ for all $i \in [n]$, for some $\delta \geq 0$. We present randomized algorithms that, for any $u \in [n]$, return an estimate $z_u$ of $z^*_u$ with additive error $\varepsilon$ or $\varepsilon \lVert z^*\rVert_\infty$, where $z^*$ is some solution to $Sz^* = b$, and the algorithm only needs to read a small portion of the input $S$ and $b$. For example, when the additive error is $\varepsilon$ and assuming $\delta>0$, we give an algorithm that runs in time $O\left( \frac{\|b\|_\infty^2 S_{\max}}{\delta^3 \varepsilon^2} \log \frac{\| b \|_\infty}{\delta \varepsilon} \right)$, where $S_{\max} = \max_{i \in [n]} |S_{ii}|$. We also prove a matching lower bound, showing that the linear dependence on $S_{\max}$ is optimal. Unlike previous sublinear-time algorithms, which apply only to symmetric diagonally dominant matrices with non-negative diagonal entries, our algorithm works for general strictly diagonally dominant matrices ($\delta > 0$) and a broader class of non-strictly diagonally dominant matrices $(\delta = 0)$. Our approach is based on analyzing a simple probabilistic recurrence satisfied by the solution. As an application, we obtain an improved sublinear-time algorithm for opinion estimation in the Friedkin--Johnsen model.
摘要：We study sublinear-time algorithms for solving linear systems $Sz = b$, where $S$ is a diagonally dominant matrix, i.e., $|S_{ii}| \geq \delta + \sum_{j \ne i} |S_{ij}|$ for all $i \in [n]$, for some $\delta \geq 0$. We present randomized algorithms that, for any $u \in [n]$, return an estimate $z_u$ of $z^*_u$ with additive error $\varepsilon$ or $\varepsilon \lVert z^*\rVert_\infty$, where $z^*$ is some solution to $Sz^* = b$, and the algorithm only needs to read a small portion of the input $S$ and $b$. For example, when the additive error is $\varepsilon$ and assuming $\delta>0$, we give an algorithm that runs in time $O\left( \frac{\|b\|_\infty^2 S_{\max}}{\delta^3 \varepsilon^2} \log \frac{\| b \|_\infty}{\delta \varepsilon} \right)$, where $S_{\max} = \max_{i \in [n]} |S_{ii}|$. We also prove a matching lower bound, showing that the linear dependence on $S_{\max}$ is optimal. Unlike previous sublinear-time algorithms, which apply only to symmetric diagonally dominant matrices with non-negative diagonal entries, our algorithm works for general strictly diagonally dominant matrices ($\delta > 0$) and a broader class of non-strictly diagonally dominant matrices $(\delta = 0)$. Our approach is based on analyzing a simple probabilistic recurrence satisfied by the solution. As an application, we obtain an improved sublinear-time algorithm for opinion estimation in the Friedkin--Johnsen model.
【5】Traces Propagation: Memory-Efficient and Scalable Forward-Only Learning in Spiking Neural Networks
标题：轨迹传播：尖峰神经网络中的内存高效且可扩展的纯前向学习
链接：https://arxiv.org/abs/2509.13053
摘要 ：尖峰神经网络（SNN）为处理动态时空信号和研究生物神经系统的学习原理提供了一个有效的框架。训练SNN的一个关键挑战是解决空间和时间的信用分配。训练SNN的主要方法是使用代理梯度的时间反向传播（BPTT）。然而，BPTT与生物神经系统中观察到的空间和时间局部性形成鲜明对比，并导致高计算和记忆需求，限制了有效的训练策略和设备上的学习。虽然现有的本地学习规则实现本地时间信用分配，利用资格痕迹，他们未能解决空间信用分配，而不诉诸辅助逐层矩阵，这增加了内存开销，阻碍了可扩展性，特别是在嵌入式设备上。在这项工作中，我们提出了痕迹传播（TP），一个只向前，内存效率，可扩展，完全本地的学习规则，结合资格痕迹与逐层对比损失，而不需要辅助逐层矩阵。TP在NMNIST和SHD数据集上的性能优于其他完全本地学习规则。在更复杂的数据集（如DVS-GESTURE和DVS-CIFAR 10）上，TP展示了具有竞争力的性能，并有效地扩展到更深的SNN架构（如VGG-9），同时与之前的完全本地可扩展规则相比，为具有大量类的数据集提供了有利的内存扩展。最后，我们证明了TP非常适合实际的微调任务，例如Google Speech Commands数据集上的关键字定位，从而为边缘的有效学习铺平了道路。
摘要：Spiking Neural Networks (SNNs) provide an efficient framework for processing dynamic spatio-temporal signals and for investigating the learning principles underlying biological neural systems. A key challenge in training SNNs is to solve both spatial and temporal credit assignment. The dominant approach for training SNNs is Backpropagation Through Time (BPTT) with surrogate gradients. However, BPTT is in stark contrast with the spatial and temporal locality observed in biological neural systems and leads to high computational and memory demands, limiting efficient training strategies and on-device learning. Although existing local learning rules achieve local temporal credit assignment by leveraging eligibility traces, they fail to address the spatial credit assignment without resorting to auxiliary layer-wise matrices, which increase memory overhead and hinder scalability, especially on embedded devices. In this work, we propose Traces Propagation (TP), a forward-only, memory-efficient, scalable, and fully local learning rule that combines eligibility traces with a layer-wise contrastive loss without requiring auxiliary layer-wise matrices. TP outperforms other fully local learning rules on NMNIST and SHD datasets. On more complex datasets such as DVS-GESTURE and DVS-CIFAR10, TP showcases competitive performance and scales effectively to deeper SNN architectures such as VGG-9, while providing favorable memory scaling compared to prior fully local scalable rules, for datasets with a significant number of classes. Finally, we show that TP is well suited for practical fine-tuning tasks, such as keyword spotting on the Google Speech Commands dataset, thus paving the way for efficient learning at the edge.
【6】Bridging Performance Gaps for Foundation Models: A Post-Training Strategy for ECGFounder
标题：弥合基础模型的绩效差距：ECGFounder的训练后策略
链接：https://arxiv.org/abs/2509.12991
备注：A simple yet effective strategy for ECG foundation models
摘要：ECG基础模型由于其在各种任务中的适应性而越来越受欢迎。然而，与特定任务模型相比，它们的临床适用性往往受到性能差距的限制，即使在大型ECG数据集上进行预训练并对目标数据进行微调之后也是如此。这种限制可能是由于缺乏有效的培训后战略。在本文中，我们提出了一种简单而有效的后训练方法来增强ECGFounder，这是一种最先进的ECG基础模型，经过超过700万次ECG记录的预训练。在PTB-XL基准测试上的实验表明，该方法在宏AUROC和宏AUPRC中分别将基线微调策略提高了1.2%-3.3%和5.3%-20.9%。此外，我们的方法优于最近的几种最先进的方法，包括特定于任务的和先进的架构。进一步的评估表明，与基线相比，我们的方法更稳定，样本效率更高，仅使用10%的训练数据，宏观AUROC就提高了9.1%，宏观AUPRC提高了34.9%。消融研究确定了有助于增强性能的关键组件，如随机深度和预览线性探测。这些发现强调了后训练策略改善ECG基础模型的潜力，我们希望这项工作将有助于ECG领域基础模型的持续发展。
摘要：ECG foundation models are increasingly popular due to their adaptability across various tasks. However, their clinical applicability is often limited by performance gaps compared to task-specific models, even after pre-training on large ECG datasets and fine-tuning on target data. This limitation is likely due to the lack of an effective post-training strategy. In this paper, we propose a simple yet effective post-training approach to enhance ECGFounder, a state-of-the-art ECG foundation model pre-trained on over 7 million ECG recordings. Experiments on the PTB-XL benchmark show that our approach improves the baseline fine-tuning strategy by 1.2%-3.3% in macro AUROC and 5.3%-20.9% in macro AUPRC. Additionally, our method outperforms several recent state-of-the-art approaches, including task-specific and advanced architectures. Further evaluation reveals that our method is more stable and sample-efficient compared to the baseline, achieving a 9.1% improvement in macro AUROC and a 34.9% improvement in macro AUPRC using just 10% of the training data. Ablation studies identify key components, such as stochastic depth and preview linear probing, that contribute to the enhanced performance. These findings underscore the potential of post-training strategies to improve ECG foundation models, and we hope this work will contribute to the continued development of foundation models in the ECG domain.
【7】Reversible Deep Equilibrium Models
标题：可逆深度均衡模型
链接：https://arxiv.org/abs/2509.12917
摘要：深度均衡模型（DEQ）是一类有趣的隐式模型，其中模型输出被隐式定义为学习函数的不动点。这些模型已经被证明在大规模任务中优于显式（固定深度）模型，通过将许多深层交换为迭代多次的单层。然而，通过DEQ的梯度计算是近似的。这通常会导致不稳定的训练动态，需要正则化或许多函数评估来修复。在这里，我们介绍了可逆深度均衡模型（RevDEQs），它允许精确的梯度计算，没有正则化，并且比DEQs少得多的函数计算。我们表明，RevDEQs在语言建模和图像分类任务上实现了最先进的性能，与可比较的隐式和显式模型相比。
摘要：Deep Equilibrium Models (DEQs) are an interesting class of implicit model where the model output is implicitly defined as the fixed point of a learned function. These models have been shown to outperform explicit (fixed-depth) models in large-scale tasks by trading many deep layers for a single layer that is iterated many times. However, gradient calculation through DEQs is approximate. This often leads to unstable training dynamics and requires regularisation or many function evaluations to fix. Here, we introduce Reversible Deep Equilibrium Models (RevDEQs) that allow for exact gradient calculation, no regularisation and far fewer function evaluations than DEQs. We show that RevDEQs achieve state-of-the-art performance on language modelling and image classification tasks against comparable implicit and explicit models.
【8】Bi-level Personalization for Federated Foundation Models: A Task-vector Aggregation Approach
标题：联邦基础模型的两级个性化：任务载体聚合方法
链接：https://arxiv.org/abs/2509.12697
摘要：联合基础模型代表了一种新的范式，可以跨客户端联合微调预先训练的基础模型。为一小群新用户或专门的场景微调基础模型仍然是一个挑战，与预训练中使用的大规模数据相比，这些场景通常涉及有限的数据。在这种情况下，个性化和联邦之间的权衡变得更加敏感。为了解决这些问题，我们提出了一个双层个性化框架，用于对基础模型进行联邦微调。具体来说，我们使用其私有数据在客户端级别上进行个性化微调，然后使用由客户端特定任务向量测量的类似用户在服务器级别上进行个性化聚合。给定从客户端级微调获得的个性化信息，服务器级个性化聚合可以获得组级个性化信息，同时减轻具有非IID数据的不相关或兴趣冲突客户端的干扰。在基准数据集上的大量实验分析证明了该算法的有效性。
摘要 ：Federated foundation models represent a new paradigm to jointly fine-tune pre-trained foundation models across clients. It is still a challenge to fine-tune foundation models for a small group of new users or specialized scenarios, which typically involve limited data compared to the large-scale data used in pre-training. In this context, the trade-off between personalization and federation becomes more sensitive. To tackle these, we proposed a bi-level personalization framework for federated fine-tuning on foundation models. Specifically, we conduct personalized fine-tuning on the client-level using its private data, and then conduct a personalized aggregation on the server-level using similar users measured by client-specific task vectors. Given the personalization information gained from client-level fine-tuning, the server-level personalized aggregation can gain group-wise personalization information while mitigating the disturbance of irrelevant or interest-conflict clients with non-IID data. The effectiveness of the proposed algorithm has been demonstrated by extensive experimental analysis in benchmark datasets.
【9】Learning to Generate Pointing Gestures in Situated Embodied Conversational Agents
标题：学习在情景对话主体中生成指向手势
链接：https://arxiv.org/abs/2509.12507
备注：DOI: 10.3389/frobt.2023.1110534. This is the author's LaTeX version
摘要：机器人技术和智能代理研究的主要目标之一是在物理环境中实现与人类的自然通信。虽然最近的工作主要集中在语言和演讲等口头模式上，但非语言沟通对于灵活的互动至关重要。我们提出了一个框架，通过模仿和强化学习相结合，在体现代理生成指向手势。使用一个小的动作捕捉数据集，我们的方法学习一个电机控制策略，产生物理上有效的，自然的姿态，具有高参考精度。我们评估的方法对监督学习和检索基线的客观指标和虚拟现实参考游戏与人类用户。结果表明，我们的系统比最先进的监督模型实现了更高的自然度和准确性，突出了模仿RL用于通信手势生成及其在机器人中的潜在应用的前景。
摘要：One of the main goals of robotics and intelligent agent research is to enable natural communication with humans in physically situated settings. While recent work has focused on verbal modes such as language and speech, non-verbal communication is crucial for flexible interaction. We present a framework for generating pointing gestures in embodied agents by combining imitation and reinforcement learning. Using a small motion capture dataset, our method learns a motor control policy that produces physically valid, naturalistic gestures with high referential accuracy. We evaluate the approach against supervised learning and retrieval baselines in both objective metrics and a virtual reality referential game with human users. Results show that our system achieves higher naturalness and accuracy than state-of-the-art supervised models, highlighting the promise of imitation-RL for communicative gesture generation and its potential application to robots.
【10】Structured Information Loss in Network Embeddings
标题：网络嵌入中的结构化信息丢失
链接：https://arxiv.org/abs/2509.12396
摘要：我们分析了一个简单的网络嵌入算法，明确描述了学习的表示完全、部分或根本不编码图的生成模型的条件。在嵌入丢失一些信息的情况下（即，是不可逆的），我们描述了等价类的图子映射到相同的嵌入，发现这些类保持社区结构，但失去了大量的密度信息。最后，我们展示了社区检测和链接预测的影响。我们的研究结果表明，仅基于嵌入的链接预测的有效性存在很大的局限性，并且我们显示了常见的条件，在这些条件下，朴素的链接预测以不成比例的方式添加边缘，可以减轻或加剧结构性偏差。
摘要：We analyze a simple algorithm for network embedding, explicitly characterizing conditions under which the learned representation encodes the graph's generative model fully, partially, or not at all. In cases where the embedding loses some information (i.e., is not invertible), we describe the equivalence classes of graphons that map to the same embedding, finding that these classes preserve community structure but lose substantial density information. Finally, we show implications for community detection and link prediction. Our results suggest strong limitations on the effectiveness of link prediction based on embeddings alone, and we show common conditions under which naive link prediction adds edges in a disproportionate manner that can either mitigate or exacerbate structural biases.
【11】FEDONet : Fourier-Embedded DeepONet for Spectrally Accurate Operator Learning
标题：FEDONet：傅里叶嵌入式DeepONet，用于光谱准确的操作员学习
链接：https://arxiv.org/abs/2509.12344
摘要：深度算子网络（DeepONets）最近成为学习非线性算子的强大数据驱动框架，特别适用于近似偏微分方程（PDE）的解决方案。尽管DeepONets的功能很有前途，但其标准实现通常在主干网络中采用完全连接的线性层，在捕获各种PDE固有的复杂空间结构时可能会遇到限制。为了解决这个问题，我们在DeepONet架构中引入了嵌入傅立叶的主干网络，利用随机傅立叶特征映射来丰富空间表示能力。我们提出的傅立叶嵌入式DeepONet，FEDONet在一系列PDE驱动的数据集上表现出优于传统DeepONet的性能，包括二维泊松方程，Burgers方程，Lorenz-63混沌系统，Eikonal方程，Allen-Cahn方程，Kuramoto-Sivashinsky方程和Lorenz-96系统。FEDONet的经验评估一致显示解决方案重建精度的显着改善，与DeepONet基线相比，平均相对L2性能增益在2- 3倍之间。这项研究强调了傅立叶嵌入在增强神经运算符学习方面的有效性，为PDE代理建模提供了一种强大且广泛适用的方法。
摘要：Deep Operator Networks (DeepONets) have recently emerged as powerful data-driven frameworks for learning nonlinear operators, particularly suited for approximating solutions to partial differential equations (PDEs). Despite their promising capabilities, the standard implementation of DeepONets, which typically employs fully connected linear layers in the trunk network, can encounter limitations in capturing complex spatial structures inherent to various PDEs. To address this, we introduce Fourier-embedded trunk networks within the DeepONet architecture, leveraging random Fourier feature mappings to enrich spatial representation capabilities. Our proposed Fourier-embedded DeepONet, FEDONet demonstrates superior performance compared to the traditional DeepONet across a comprehensive suite of PDE-driven datasets, including the two-dimensional Poisson equation, Burgers' equation, the Lorenz-63 chaotic system, Eikonal equation, Allen-Cahn equation, Kuramoto-Sivashinsky equation, and the Lorenz-96 system. Empirical evaluations of FEDONet consistently show significant improvements in solution reconstruction accuracy, with average relative L2 performance gains ranging between 2-3x compared to the DeepONet baseline. This study highlights the effectiveness of Fourier embeddings in enhancing neural operator learning, offering a robust and broadly applicable methodology for PDE surrogate modeling.
【12】AIssistant: An Agentic Approach for Human--AI Collaborative Scientific Work on Reviews and Perspectives in Machine Learning
标题：Aissistant：人类与人工智能的一种统计方法关于机器学习评论和观点的协作科学工作
链接：https://arxiv.org/abs/2509.12282
摘要：人工智能辅助研究的进展为文献检索、假设生成、实验和手稿准备引入了强大的工具。然而，各系统仍然支离破碎，缺乏以人为本的工作流程。为了解决这些差距，我们引入了AIssistant，这是一个开源的人机协作框架，旨在简化科学工作流程的端到端创建。由于我们的开发仍处于早期阶段，我们在这里介绍了AIssistant的第一个实验，以透视和回顾机器学习的研究论文。我们的系统集成了模块化工具和代理，用于文献综合，分段实验，引文管理和自动LaTeX论文文本生成，同时在每个阶段保持人为监督，以确保准确性，连贯性和学术严谨性。我们在三个层面进行了全面评估：（1）独立的人工审查，遵循NeurIPS双盲标准;（2）自动LLM审查，使用GPT-5作为可扩展的人工审查代理;（3）程序主席监督，主席监督整个审查过程并做出最终验证和验收决定。实验结果表明，AIssistant提高了写作效率和主题一致性。尽管如此，人类与人工智能的合作对于保持事实的正确性、方法的合理性和道德合规性仍然至关重要。尽管它的有效性，我们确定的关键限制，包括幻觉引用，难以适应动态的论文结构，和不完整的多模态内容的整合。
摘要 ：Advances in AI-assisted research have introduced powerful tools for literature retrieval, hypothesis generation, experimentation, and manuscript preparation. However, systems remain fragmented and lack human-centred workflows. To address these gaps, we introduce AIssistant, an agentic, open-source Human-AI collaborative framework designed to simplify the end-to-end creation of scientific workflows. Since our development is still in an early stage, we present here the first experiments with AIssistant for perspective and review research papers in machine learning. Our system integrates modular tools and agents for literature synthesis, section-wise experimentation, citation management, and automatic LaTeX paper text generation, while maintaining human oversight at every stage to ensure accuracy, coherence, and scholarly rigour. We conducted a comprehensive evaluation across three layers: (1) Independent Human Review, following NeurIPS double-blind standards; (2) Automated LLM Review, using GPT-5 as a scalable human review proxy; and (3) Program Chair Oversight, where the chair monitors the entire review process and makes final validation and acceptance decisions. The results demonstrate that AIssistant improves drafting efficiency and thematic consistency. Nonetheless, Human-AI collaboration remains essential for maintaining factual correctness, methodological soundness, and ethical compliance. Despite its effectiveness, we identify key limitations, including hallucinated citations, difficulty adapting to dynamic paper structures, and incomplete integration of multimodal content.
【13】Scaling Up Data Parallelism in Decentralized Deep Learning
标题：扩大去中心化深度学习中的数据并行主义
链接：https://arxiv.org/abs/2509.12213
摘要：尽管在理论上已经进行了广泛的探索，但分散式学习尚未被生产使用，这主要是由于大规模DNN训练缺乏稳定性，可扩展性和通用性。为了阐明分散学习的生产使用，这项工作研究了大规模的分散数据并行训练。为此，我们引入了一个基准测试框架，即DBench，来托管集中式和分散式DNN培训。在DBench的基础上，我们引入了一种基准测试方法，通过改变通信图和训练尺度来揭示模型精度与参数张量方差之间的相关性。基于基准测试结果，我们观察到：（1）与集中式学习类似，分散式数据并行训练在训练规模扩大时也存在可扩展性和通用性问题;（2）分散式学习的模型精度与通信图中的连接数相关;（3）分散学习的模型精度对模型副本中参数张量的方差非常敏感。基于这些观察，我们提出了Ada，这是一种分散的自适应方法，它遵循分散的SGD方法进行大规模DNN训练，并在整个训练迭代过程中动态地调整使用的通信图。我们将Ada应用于大规模训练，并观察到Ada可以在分散式DNN训练中始终获得最佳收敛率，并为所有样本应用程序提供与集中式学习相同或更好的模型精度，即使在1008 GPU的规模上为ImageNet-1 K训练ResNet 50时也是如此。
摘要：Although it has been extensively explored in theory, decentralized learning is not yet green-lighted for production use, largely due to a lack of stability, scalability, and generality in large scale DNN training. To shed light on the production use of decentralized learning, this work studies decentralized data parallel training at scale. To this end, we introduce a benchmarking framework, namely DBench, to host both centralized and decentralized DNN training. Building upon DBench, we introduce a benchmarking methodology to uncover the correlations between model accuracy and the variances of parameter tensors by varying communication graphs and training scales. Based on the benchmarking results, we observe that, (1) Similar to centralized learning, decentralized data parallel training also presents the issues of scalability and generality when the training scales up; (2) The model accuracy of decentralized learning is correlated to the number of connections in a communication graph; (3) The model accuracy of decentralized learning is surprisingly sensitive to the variance of parameter tensors across model replicas. Built upon the observations, we propose Ada, a decentralized adaptive approach that performs large scale DNN training following a decentralized SGD method and adapting the communication graph in use dynamically throughout training iterations. We apply Ada on large scale training and observe that Ada can obtain the best convergence rates consistently in decentralized DNN training, and delivers equally or comparably good model accuracy for all sample applications as centralized learning does, even when training ResNet50 for ImageNet-1K on the scale of 1008 GPUs.
【14】Fast reconstruction of degenerate populations of conductance-based neuron models from spike times
标题：从峰值时间快速重建基于电导的神经元模型的简并种群
链接：https://arxiv.org/abs/2509.12783
摘要：神经元通过尖峰信号进行通信，而尖峰信号的定时是神经元处理的关键部分。尖峰时间可以在细胞内和细胞外通过实验记录，并且是最先进的神经探针的主要输出。另一方面，神经元的活动在分子水平上由许多不同的跨膜蛋白（称为离子通道）产生的电流控制。迄今为止，将尖峰时间与离子通道组成相联系仍然是一项艰巨的任务。为了应对这一挑战，我们开发了一种方法，将深度学习与称为动态输入电导（DIC）的理论工具相结合，将离子通道相互作用的复杂性降低为三个可解释的组件，描述神经元如何尖峰。我们的方法使用深度学习直接从尖峰时间推断DIC，然后生成“孪生”神经元模型的群体，这些模型复制观察到的活动，同时捕获膜通道组成的自然变异性。该方法快速、准确，并且仅使用尖峰记录。我们还提供带有图形界面的开源软件，使没有编程专业知识的研究人员也可以使用。
摘要：Neurons communicate through spikes, and spike timing is a crucial part of neuronal processing. Spike times can be recorded experimentally both intracellularly and extracellularly, and are the main output of state-of-the-art neural probes. On the other hand, neuronal activity is controlled at the molecular level by the currents generated by many different transmembrane proteins called ion channels. Connecting spike timing to ion channel composition remains an arduous task to date. To address this challenge, we developed a method that combines deep learning with a theoretical tool called Dynamic Input Conductances (DICs), which reduce the complexity of ion channel interactions into three interpretable components describing how neurons spike. Our approach uses deep learning to infer DICs directly from spike times and then generates populations of "twin" neuron models that replicate the observed activity while capturing natural variability in membrane channel composition. The method is fast, accurate, and works using only spike recordings. We also provide open-source software with a graphical interface, making it accessible to researchers without programming expertise.
【15】PBPK-iPINNs : Inverse Physics-Informed Neural Networks for Physiologically Based Pharmacokinetic Brain Models
标题：PBPK-iPINN：基于生理的药代动力学大脑模型的逆物理信息神经网络
链接：https://arxiv.org/abs/2509.12666
备注：24 pages, 11 figures
摘要：物理信息神经网络（PINN）利用机器学习和微分方程来解决正问题和逆问题，确保预测遵循物理定律。基于生理学的药代动力学（PBPK）建模通过使用以生理学为重点的机制框架而超越了经典房室方法。PBPK模型基于ODE系统，其中每个方程表示隔室（例如器官或组织）中药物的质量平衡。这些ODE包括反映生理、生化和药物特异性特征的参数，以模拟药物如何在体内移动。在本文中，我们介绍了PBPK-iPINN，一种方法来估计药物特异性或患者特异性参数和药物浓度分布在PBPK脑室模型使用逆PINN。我们证明，逆问题收敛到正确的解决方案，损失函数组件（数据丢失，初始条件损失和剩余损失）必须适当加权，参数（包括层数，神经元数量，激活函数，学习率，优化器和配置点）必须仔细调整。PBPK-iPINN方法的性能进行比较，建立传统的数值和统计方法。
摘要：Physics-Informed Neural Networks (PINNs) leverage machine learning with differential equations to solve direct and inverse problems, ensuring predictions follow physical laws. Physiologically based pharmacokinetic (PBPK) modeling advances beyond classical compartmental approaches by using a mechanistic, physiology focused framework. A PBPK model is based on a system of ODEs, with each equation representing the mass balance of a drug in a compartment, such as an organ or tissue. These ODEs include parameters that reflect physiological, biochemical, and drug-specific characteristics to simulate how the drug moves through the body. In this paper, we introduce PBPK-iPINN, a method to estimate drug-specific or patient-specific parameters and drug concentration profiles in PBPK brain compartment models using inverse PINNs. We demonstrate that, for the inverse problem to converge to the correct solution, the loss function components (data loss, initial conditions loss, and residual loss) must be appropriately weighted, and parameters (including number of layers, number of neurons, activation functions, learning rate, optimizer, and collocation points) must be carefully tuned. The performance of the PBPK-iPINN approach is then compared with established traditional numerical and statistical methods.
【16】Genome-Factory: An Integrated Library for Tuning, Deploying, and Interpreting Genomic Models
标题：基因组工厂：用于调整、部署和解释基因组模型的集成库
链接：https://arxiv.org/abs/2509.12266
摘要：我们介绍Genome-Factory，一个集成的Python库，用于调整，部署和解释基因组模型。我们的核心贡献是简化和统一基因组模型开发的工作流程：数据收集，模型调整，推理，基准测试和可解释性。对于数据收集，Genome-Factory提供了一个自动化的管道来下载基因组序列并对其进行预处理。它还包括质量控制，如GC含量标准化。对于模型调整，Genome-Factory支持三种方法：全参数，低秩自适应和基于适配器的微调。它与广泛的基因组模型兼容。对于推理，Genome-Factory支持嵌入提取和DNA序列生成。对于基准测试，我们包括两个现有的基准测试，并为用户提供一个灵活的界面，以纳入其他基准测试。对于可解释性，Genome-Factory引入了第一个基于稀疏自动编码器的开源生物解释器。该模块将嵌入分解为稀疏的、接近单语义的潜在单元，并通过回归外部读数将它们与可解释的基因组特征联系起来。为了提高可访问性，Genome-Factory具有零代码命令行界面和用户友好的Web界面。我们在三个维度上验证了基因组工厂的实用性：（i）与不同模型和微调方法的兼容性;（ii）使用两个开源基准对下游性能进行基准测试;（iii）使用DNABERT-2对学习表示进行生物学解释。这些结果突出了其端到端的可用性和实际价值，为现实世界的基因组分析。
摘要：We introduce Genome-Factory, an integrated Python library for tuning, deploying, and interpreting genomic models. Our core contribution is to simplify and unify the workflow for genomic model development: data collection, model tuning, inference, benchmarking, and interpretability. For data collection, Genome-Factory offers an automated pipeline to download genomic sequences and preprocess them. It also includes quality control, such as GC content normalization. For model tuning, Genome-Factory supports three approaches: full-parameter, low-rank adaptation, and adapter-based fine-tuning. It is compatible with a wide range of genomic models. For inference, Genome-Factory enables both embedding extraction and DNA sequence generation. For benchmarking, we include two existing benchmarks and provide a flexible interface for users to incorporate additional benchmarks. For interpretability, Genome-Factory introduces the first open-source biological interpreter based on a sparse auto-encoder. This module disentangles embeddings into sparse, near-monosemantic latent units and links them to interpretable genomic features by regressing on external readouts. To improve accessibility, Genome-Factory features both a zero-code command-line interface and a user-friendly web interface. We validate the utility of Genome-Factory across three dimensions: (i) Compatibility with diverse models and fine-tuning methods; (ii) Benchmarking downstream performance using two open-source benchmarks; (iii) Biological interpretation of learned representations with DNABERT-2. These results highlight its end-to-end usability and practical value for real-world genomic analysis.
【17】Physics-Informed Neural Networks vs. Physics Models for Non-Invasive Glucose Monitoring: A Comparative Study Under Realistic Synthetic Conditions
标题：无创血糖监测的物理信息神经网络与物理模型：现实合成条件下的比较研究
链接：https://arxiv.org/abs/2509.12253
摘要：非侵入式血糖监测仪在实验室外经常失败，因为现有的数据集忽略了硬件噪声，环境漂移和人与人之间的生理学。我们推出了第一款超逼真的近红外（NIR）模拟器，可注入12位ADC量化、+/-0.1% LED老化、光电二极管暗噪声、15-45 C温度、30-90%相对湿度、接触压力变化、Fitzpatrick I-VI黑色素和昼夜血糖波动（黎明现象）。使用这个平台（rho glucose-NIR = 0.21），我们对六种方法进行了基准测试：增强型Beer-Lambert（物理工程岭回归），三种物理信息神经网络（PINN），选择性辐射转移PINN和浅DNN。Beer-Lambert仅使用56个参数和0.01 ms推断就实现了13.6 mg/dL RMSE、95.8% Clarke-A和93.8% +/-15%准确度，优于最佳PINN（14.6 mg/dL）和SDNN基线（35.1 mg/dL）。结果推翻了更深的PINN占主导地位的假设，并为嵌入式光学葡萄糖传感器的快速原型设计提供了开放的端到端参考堆栈。
摘要：Non-invasive glucose monitors often fail outside the lab because existing datasets ignore hardware noise, environmental drift, and person-to-person physiology. We introduce the first ultra-realistic near-infrared (NIR) simulator that injects 12-bit ADC quantisation, +/-0.1% LED ageing, photodiode dark noise, 15-45 C temperature, 30-90% relative humidity, contact-pressure variation, Fitzpatrick I-VI melanin, and diurnal glucose excursions (dawn phenomenon). Using this platform (rho glucose-NIR = 0.21), we benchmark six methods: Enhanced Beer-Lambert (physics-engineered ridge regression), three physics-informed neural networks (PINNs), a selective radiative-transfer PINN, and a shallow DNN. Beer-Lambert achieves 13.6 mg/dL RMSE, 95.8% Clarke-A and 93.8% +/-15% accuracy with only 56 parameters and 0.01 ms inference, outperforming the best PINN (14.6 mg/dL) and the SDNN baseline (35.1 mg/dL). Results overturn the assumption that deeper PINNs dominate and supply an open, end-to-end reference stack for rapid prototyping of embedded optical glucose sensors.
其他(33篇)
【1】Intelligent Vacuum Thermoforming Process
标题：智能真空热成型工艺
链接：https://arxiv.org/abs/2509.13250
备注：Contains 6 figures in total, 15 pages. Under revision for Journal of Intelligent Manufacturing
摘要：由于材料特性和模具配置的变化，确保真空热成型的质量一致性面临挑战。本研究引入了一种基于视觉的质量控制系统来预测和优化工艺参数，从而以最少的数据要求提高零件质量。一个全面的数据集开发使用视觉数据从真空成型的样品进行各种工艺参数，辅以图像增强技术，以提高模型训练。随后采用k-最近邻算法通过将低质量零件映射到高质量零件来确定工艺参数所需的调整。该模型在调节加热功率、加热时间、抽真空时间等方面表现出较强的性能，减少了缺陷，提高了生产效率。
摘要：Ensuring consistent quality in vacuum thermoforming presents challenges due to variations in material properties and tooling configurations. This research introduces a vision-based quality control system to predict and optimise process parameters, thereby enhancing part quality with minimal data requirements. A comprehensive dataset was developed using visual data from vacuum-formed samples subjected to various process parameters, supplemented by image augmentation techniques to improve model training. A k-Nearest Neighbour algorithm was subsequently employed to identify adjustments needed in process parameters by mapping low-quality parts to their high-quality counterparts. The model exhibited strong performance in adjusting heating power, heating time, and vacuum time to reduce defects and improve production efficiency.
【2】Don't Forget the Nonlinearity: Unlocking Activation Functions in Efficient Fine-Tuning
标题：不要忘记非线性：在高效微调中解锁激活功能
链接：https://arxiv.org/abs/2509.13240
摘要：现有的参数有效的微调（PEFT）方法主要是适应权重矩阵，同时保持激活函数固定。我们引入了\textbf{NoRA}，这是第一个直接在预训练的基于transformer的模型中适应非线性激活函数的PEFT框架。NoRA用可学习的有理函数取代固定激活，并将结构化的低秩更新应用于分子和分母系数，并采用分组设计，以最小的成本定位自适应并提高稳定性。在CIFAR-10和CIFAR-100上训练的Vision Transformers上，NoRA匹配或超过完全微调，同时仅更新0.4\%的参数（0.02M），实现了+0.17\%和+0.27\%的精度增益。当与LoRA（\textbf{NoRA++}）相结合时，它通过添加更少的可训练参数，在匹配的训练预算下优于LoRA和DoRA。在LLaMA 3 -8B指令调优上，NoRA++持续提高生成质量，平均MMLU增益为+0.3\%--0.8\%，包括STEM（Alpaca）上的+1.6\%和OpenOrca上的+1.3\%。我们进一步表明，NoRA约束适应低维函数子空间，隐式正则化更新幅度和方向。这些结果建立激活空间调整作为一个补充和高度参数有效的替代基于权重的PEFT，定位激活功能作为第一类对象的模型自适应。
摘要 ：Existing parameter-efficient fine-tuning (PEFT) methods primarily adapt weight matrices while keeping activation functions fixed. We introduce \textbf{NoRA}, the first PEFT framework that directly adapts nonlinear activation functions in pretrained transformer-based models. NoRA replaces fixed activations with learnable rational functions and applies structured low-rank updates to numerator and denominator coefficients, with a group-wise design that localizes adaptation and improves stability at minimal cost. On vision transformers trained on CIFAR-10 and CIFAR-100, NoRA matches or exceeds full fine-tuning while updating only 0.4\% of parameters (0.02M), achieving accuracy gains of +0.17\% and +0.27\%. When combined with LoRA (\textbf{NoRA++}), it outperforms LoRA and DoRA under matched training budgets by adding fewer trainable parameters. On LLaMA3-8B instruction tuning, NoRA++ consistently improves generation quality, yielding average MMLU gains of +0.3\%--0.8\%, including +1.6\% on STEM (Alpaca) and +1.3\% on OpenOrca. We further show that NoRA constrains adaptation to a low-dimensional functional subspace, implicitly regularizing update magnitude and direction. These results establish activation-space tuning as a complementary and highly parameter-efficient alternative to weight-based PEFT, positioning activation functions as first-class objects for model adaptation.
【3】Density-Aware Farthest Point Sampling
标题：密度感知法拉第点采样
链接：https://arxiv.org/abs/2509.13213
备注：12 pages, 2 figures
摘要：我们专注于在由于计算约束或高标记成本而限制标记训练数据可用性的情况下训练机器学习回归模型。因此，从未标记数据中选择合适的训练集对于平衡性能和效率至关重要。对于训练数据的选择，我们专注于被动和模型无关的采样方法，只考虑数据特征表示。我们推导出Lipschitz连续回归模型的预期预测误差的上界，该上界线性地依赖于训练集的加权填充距离，我们可以通过考虑数据特征来简单地估计这个量。本文介绍了一种新的采样方法--密度感知Faradian点采样（DA-FPS）。我们证明了DA-FPS提供了近似极小值的数据驱动的估计加权填充距离，从而旨在最大限度地减少我们的推导出的界限。我们在三个数据集上使用两个回归模型进行实验。结果表明，DA-FPS显着降低了平均绝对预测误差相比，其他采样策略。
摘要：We focus on training machine learning regression models in scenarios where the availability of labeled training data is limited due to computational constraints or high labeling costs. Thus, selecting suitable training sets from unlabeled data is essential for balancing performance and efficiency. For the selection of the training data, we focus on passive and model-agnostic sampling methods that only consider the data feature representations. We derive an upper bound for the expected prediction error of Lipschitz continuous regression models that linearly depends on the weighted fill distance of the training set, a quantity we can estimate simply by considering the data features. We introduce "Density-Aware Farthest Point Sampling" (DA-FPS), a novel sampling method. We prove that DA-FPS provides approximate minimizers for a data-driven estimation of the weighted fill distance, thereby aiming at minimizing our derived bound. We conduct experiments using two regression models across three datasets. The results demonstrate that DA-FPS significantly reduces the mean absolute prediction error compared to other sampling strategies.
【4】Concentration inequalities for semidefinite least squares based on data
标题：基于数据的半定最小二乘集中不等式
链接：https://arxiv.org/abs/2509.13166
摘要：我们研究数据驱动的最小二乘（LS）问题的半定（SD）的约束，并推导出有限样本的保证，当这些约束放松时，他们的最优解的频谱。特别是，我们提供了一个高置信度的界限，允许一个更简单的程序来代替完整的SDLS问题，同时确保所得到的解决方案的特征值是$\varepsilon$-关闭的SD约束所执行的。所开发的证书随着数据数量的增加而不断缩小，结果证明易于计算，无分布，并且只需要独立和同分布的样本。此外，当SDLS是用来学习一个未知的二次函数，我们建立的梯度下降最小化代理成本之间的误差界限，没有SD约束和真正的最小值。
摘要：We study data-driven least squares (LS) problems with semidefinite (SD) constraints and derive finite-sample guarantees on the spectrum of their optimal solutions when these constraints are relaxed. In particular, we provide a high confidence bound allowing one to solve a simpler program in place of the full SDLS problem, while ensuring that the eigenvalues of the resulting solution are $\varepsilon$-close of those enforced by the SD constraints. The developed certificate, which consistently shrinks as the number of data increases, turns out to be easy-to-compute, distribution-free, and only requires independent and identically distributed samples. Moreover, when the SDLS is used to learn an unknown quadratic function, we establish bounds on the error between a gradient descent iterate minimizing the surrogate cost obtained with no SD constraints and the true minimizer.
【5】When Inverse Data Outperforms: Exploring the Pitfalls of Mixed Data in Multi-Stage Fine-Tuning
标题：当反向数据表现出色时：探索多阶段微调中混合数据的陷阱
链接：https://arxiv.org/abs/2509.13079
摘要：现有的工作表明，o 1级的性能可以实现有限的数据蒸馏，但大多数现有的方法侧重于单向监督微调（SFT），忽略了不同的推理模式之间错综复杂的相互作用。在本文中，我们构建了r1 k，这是一个高质量的反向推理数据集，通过从s1 k中反演1，000个正向示例来获得，并研究了SFT和直接偏好优化（DPO）如何影响双向推理目标下的对齐。在评估的基准测试中，r1 k上的SFT比s1 k的精度提高了1.6%-6.8%。然而，在SFT期间天真地混合正向和反向数据会削弱方向区分。尽管DPO可以部分恢复这种区别，但它也通过将概率质量转移到不相关的输出来抑制不太首选的推理路径。这些发现表明，混合推理数据引入相互冲突的监督信号，强调需要强大的和方向感知的对齐策略。
摘要：Existing work has shown that o1-level performance can be achieved with limited data distillation, but most existing methods focus on unidirectional supervised fine-tuning (SFT), overlooking the intricate interplay between diverse reasoning patterns. In this paper, we construct r1k, a high-quality reverse reasoning dataset derived by inverting 1,000 forward examples from s1k, and examine how SFT and Direct Preference Optimization (DPO) affect alignment under bidirectional reasoning objectives. SFT on r1k yields a 1.6%--6.8% accuracy improvement over s1k across evaluated benchmarks. However, naively mixing forward and reverse data during SFT weakens the directional distinction. Although DPO can partially recover this distinction, it also suppresses less preferred reasoning paths by shifting the probability mass toward irrelevant outputs. These findings suggest that mixed reasoning data introduce conflicting supervision signals, underscoring the need for robust and direction-aware alignment strategies.
【6】Spiking Vocos: An Energy-Efficient Neural Vocoder
标题：Spiking Vocos：一种节能的神经声码器
链接：https://arxiv.org/abs/2509.13049
摘要 ：尽管神经声码器在合成速度和保真度方面取得了显着进步，但其高能耗仍然是在计算受限的边缘设备上实际部署的关键障碍。尖峰神经网络（SNN）因其事件驱动的性质而被广泛认可的高能效，为低资源场景提供了一个有前途的解决方案。在本文中，我们提出了尖峰Vocos，一种新的尖峰神经声码器，具有超低能耗，建立在有效的Vocos框架。为了缓解SNN中固有的信息瓶颈，我们设计了一个Spiking ConvNeXt模块，以减少乘法累加（MAC）操作，并结合幅度捷径路径来保持关键的信号动态。此外，为了弥补与人工神经网络（ANN）的性能差距，我们引入了一种自架构蒸馏策略来有效地传递知识。一个轻量级的时间转移模块也被集成，以提高模型的能力，融合跨时间维度的信息，可以忽略不计的计算开销。实验表明，我们的模型实现了性能相媲美的人工神经网络，UTMOS和PESQ得分分别为3.74和3.45，而消耗的能量只有14.7%。源代码可在https://github.com/pymaster17/Spiking-Vocos上获得。
摘要：Despite the remarkable progress in the synthesis speed and fidelity of neural vocoders, their high energy consumption remains a critical barrier to practical deployment on computationally restricted edge devices. Spiking Neural Networks (SNNs), widely recognized for their high energy efficiency due to their event-driven nature, offer a promising solution for low-resource scenarios. In this paper, we propose Spiking Vocos, a novel spiking neural vocoder with ultra-low energy consumption, built upon the efficient Vocos framework. To mitigate the inherent information bottleneck in SNNs, we design a Spiking ConvNeXt module to reduce Multiply-Accumulate (MAC) operations and incorporate an amplitude shortcut path to preserve crucial signal dynamics. Furthermore, to bridge the performance gap with its Artificial Neural Network (ANN) counterpart, we introduce a self-architectural distillation strategy to effectively transfer knowledge. A lightweight Temporal Shift Module is also integrated to enhance the model's ability to fuse information across the temporal dimension with negligible computational overhead. Experiments demonstrate that our model achieves performance comparable to its ANN counterpart, with UTMOS and PESQ scores of 3.74 and 3.45 respectively, while consuming only 14.7% of the energy. The source code is available at https://github.com/pymaster17/Spiking-Vocos.
【7】Data-driven Methods of Extracting Text Structure and Information Transfer
标题：提取文本结构和信息传输的数据驱动方法
链接：https://arxiv.org/abs/2509.12999
摘要：安娜·卡列尼娜原则（Anna Karenina Principle，AKP）认为，成功需要满足一小部分基本条件，而失败则有多种形式。我们测试AKP，它的反向，和两个进一步的模式描述为有序和嘈杂的小说，在线百科全书，研究论文和电影。文本被表示为功能块的序列，并在过渡顺序和位置的收敛性进行评估。结果表明，结构原则不同的媒体：小说遵循反向AKP顺序，维基百科结合AKP与有序模式，学术论文显示反向AKP顺序，但仍然嘈杂的位置，和电影分歧的体裁。因此，成功取决于特定于每种媒介的结构性限制，而失败则在不同领域呈现不同的形式。
摘要：The Anna Karenina Principle (AKP) holds that success requires satisfying a small set of essential conditions, whereas failure takes diverse forms. We test AKP, its reverse, and two further patterns described as ordered and noisy across novels, online encyclopedias, research papers, and movies. Texts are represented as sequences of functional blocks, and convergence is assessed in transition order and position. Results show that structural principles vary by medium: novels follow reverse AKP in order, Wikipedia combines AKP with ordered patterns, academic papers display reverse AKP in order but remain noisy in position, and movies diverge by genre. Success therefore depends on structural constraints that are specific to each medium, while failure assumes different shapes across domains.
【8】Causal Discovery via Quantile Partial Effect
标题：通过分位数部分效应发现原因
链接：https://arxiv.org/abs/2509.12981
备注：29 pages, 6 figures
摘要：分位数部分效应（Quantile Partial Effect，QPE）是与条件分位数回归相关的统计量，用于测量协变量在不同水平的效应。我们的理论表明，当因果关系的QPE被假定为位于一个有限的线性跨度，原因和效果是可识别的，从它们的观测分布。这概括了以前的可识别性结果的基础上的功能因果模型（FCM）与添加剂，异方差噪声等，同时，由于QPE完全驻留在观测水平，这个参数假设不需要考虑机制，噪声，甚至马尔可夫假设，而是直接利用形状特征的不对称性在观测分布。通过对估计的QPE执行基函数测试，可以区分因果方向，这在大量双变量因果发现数据集的实验中根据经验证明是有效的。对于多变量因果发现，利用QPE和得分函数之间的密切联系，我们发现，Fisher信息是足够的统计措施，以确定因果顺序时，对QPE的二阶矩的假设。我们验证了使用Fisher信息来识别多个合成和真实世界的多变量因果发现数据集上的因果顺序的可行性。
摘要：Quantile Partial Effect (QPE) is a statistic associated with conditional quantile regression, measuring the effect of covariates at different levels. Our theory demonstrates that when the QPE of cause on effect is assumed to lie in a finite linear span, cause and effect are identifiable from their observational distribution. This generalizes previous identifiability results based on Functional Causal Models (FCMs) with additive, heteroscedastic noise, etc. Meanwhile, since QPE resides entirely at the observational level, this parametric assumption does not require considering mechanisms, noise, or even the Markov assumption, but rather directly utilizes the asymmetry of shape characteristics in the observational distribution. By performing basis function tests on the estimated QPE, causal directions can be distinguished, which is empirically shown to be effective in experiments on a large number of bivariate causal discovery datasets. For multivariate causal discovery, leveraging the close connection between QPE and score functions, we find that Fisher Information is sufficient as a statistical measure to determine causal order when assumptions are made about the second moment of QPE. We validate the feasibility of using Fisher Information to identify causal order on multiple synthetic and real-world multivariate causal discovery datasets.
【9】MMMS: Multi-Modal Multi-Surface Interactive Segmentation
标题：MMMS：多模式多表面交互分割
链接：https://arxiv.org/abs/2509.12963
备注：19 pages, 11 figures, 10 pages
摘要：在本文中，我们提出了一种方法来交互式创建分割掩模的基础上，用户点击。我们特别注意同时存在于同一图像中的多个表面的分割。由于这些表面可能是严重纠缠和相邻的，我们还提出了一种新的扩展评估指标，占这种情况下的挑战。此外，所提出的方法能够使用多模态输入来促进分割任务。该方法的核心是一个网络架构，它将RGB图像、许多非RGB模态、错误掩码和编码点击作为输入。基于此输入，网络预测改进的分割掩码。我们设计我们的架构，使其符合两个条件：（1）RGB骨干只能作为一个黑盒。(2)为了减少响应时间，我们希望我们的模型在图像特征提取和多模态融合之后集成特定于交互的信息。我们将整个任务称为多模态多表面交互式分割（MMMS）。我们能够证明我们的多模态融合策略的有效性。使用其他模态，我们的系统在DeLiVER上将NoC@90平均每个表面减少多达1.28次点击，在MFNet上减少多达1.19次点击。最重要的是，我们能够证明，我们的仅RGB基线在经典的单掩模交互式分割场景中进行测试时具有竞争力，在某些情况下甚至具有卓越的性能。
摘要：In this paper, we present a method to interactively create segmentation masks on the basis of user clicks. We pay particular attention to the segmentation of multiple surfaces that are simultaneously present in the same image. Since these surfaces may be heavily entangled and adjacent, we also present a novel extended evaluation metric that accounts for the challenges of this scenario. Additionally, the presented method is able to use multi-modal inputs to facilitate the segmentation task. At the center of this method is a network architecture which takes as input an RGB image, a number of non-RGB modalities, an erroneous mask, and encoded clicks. Based on this input, the network predicts an improved segmentation mask. We design our architecture such that it adheres to two conditions: (1) The RGB backbone is only available as a black-box. (2) To reduce the response time, we want our model to integrate the interaction-specific information after the image feature extraction and the multi-modal fusion. We refer to the overall task as Multi-Modal Multi-Surface interactive segmentation (MMMS). We are able to show the effectiveness of our multi-modal fusion strategy. Using additional modalities, our system reduces the NoC@90 by up to 1.28 clicks per surface on average on DeLiVER and up to 1.19 on MFNet. On top of this, we are able to show that our RGB-only baseline achieves competitive, and in some cases even superior performance when tested in a classical, single-mask interactive segmentation scenario.
【10】Rethinking the Evaluation of Alignment Methods: Insights into Diversity, Generalisation, and Safety
标题：重新思考对齐方法的评估：对多样性，泛化和安全性的见解
链接：https://arxiv.org/abs/2509.12936
摘要：大型语言模型（LLM）需要仔细调整以平衡相互竞争的目标-真实性，安全性，简洁性，主动性和多样性。现有的研究集中在个别技术或具体的方面，缺乏一个整体的评估固有的权衡。我们提出了一个统一的评估框架，比较LLM对齐方法（PPO，DPO，ORPO，KTO）在这五个轴上，使用分布和分布数据集。利用专门的法学硕士作为法官提示，通过人体研究验证，我们发现DPO和KTO在事实准确性方面表现出色，PPO和DPO在安全性方面领先，PPO最好地平衡了简洁性与主动性。我们的研究结果提供了对常见对齐方法的权衡的见解，指导更平衡和可靠的LLM的开发。
摘要：Large language models (LLMs) require careful alignment to balance competing objectives - factuality, safety, conciseness, proactivity, and diversity. Existing studies focus on individual techniques or specific dimensions, lacking a holistic assessment of the inherent trade-offs. We propose a unified evaluation framework that compares LLM alignment methods (PPO, DPO, ORPO, KTO) across these five axes, using both in-distribution and out-of-distribution datasets. Leveraging a specialized LLM-as-Judge prompt, validated through human studies, we reveal that DPO and KTO excel in factual accuracy, PPO and DPO lead in safety, and PPO best balances conciseness with proactivity. Our findings provide insights into trade-offs of common alignment methods, guiding the development of more balanced and reliable LLMs.
【11】HLSMAC: A New StarCraft Multi-Agent Challenge for High-Level Strategic Decision-Making
标题：HLSMAC：针对高级战略决策的新星际争霸多智能体挑战
链接：https://arxiv.org/abs/2509.12927
备注：30 pages, 13 figures with appendix
摘要：基准测试对于评估多智能体强化学习（MARL）算法至关重要。虽然《星际争霸2》相关的环境推动了MARL的重大进步，但现有的基准（如SMAC）主要集中在微观管理上，限制了对高级战略情报的全面评估。为了解决这个问题，我们引入HLSMAC，一个新的合作MARL基准与12个精心设计的星际争霸II场景的基础上，从36战略的经典战略。每个场景都对应于一个特定的战略，旨在挑战具有不同战略要素的智能体，包括战术机动、时间协调和欺骗，从而为评估高级别战略决策能力开辟了途径。我们还提出了新的指标，在多个维度超越传统的胜率，如能力利用率和推进效率，评估代理的整体表现在HLSMAC环境。我们将最先进的MARL算法和基于LLM的代理与我们的基准集成在一起，并进行全面的实验。结果表明，HLSMAC作为一个强大的测试平台，推进多智能体战略决策。
摘要：Benchmarks are crucial for assessing multi-agent reinforcement learning (MARL) algorithms. While StarCraft II-related environments have driven significant advances in MARL, existing benchmarks like SMAC focus primarily on micromanagement, limiting comprehensive evaluation of high-level strategic intelligence. To address this, we introduce HLSMAC, a new cooperative MARL benchmark with 12 carefully designed StarCraft II scenarios based on classical stratagems from the Thirty-Six Stratagems. Each scenario corresponds to a specific stratagem and is designed to challenge agents with diverse strategic elements, including tactical maneuvering, timing coordination, and deception, thereby opening up avenues for evaluating high-level strategic decision-making capabilities. We also propose novel metrics across multiple dimensions beyond conventional win rate, such as ability utilization and advancement efficiency, to assess agents' overall performance within the HLSMAC environment. We integrate state-of-the-art MARL algorithms and LLM-based agents with our benchmark and conduct comprehensive experiments. The results demonstrate that HLSMAC serves as a robust testbed for advancing multi-agent strategic decision-making.
【12】Soft Gradient Boosting with Learnable Feature Transforms for Sequential Regression
标题：利用可学习特征变换进行软梯度提升以实现序列回归
链接：https://arxiv.org/abs/2509.12920
摘要：我们提出了一个软梯度提升框架，用于顺序回归，该框架在提升过程中嵌入了一个可学习的线性特征变换。在每次提升迭代中，我们训练软决策树并一起学习线性输入特征变换Q。这种方法在高维、数据稀缺的场景中特别有利，因为它在提升的同时发现了最相关的输入表示。我们使用合成和真实世界的数据集证明，我们的方法通过特征选择/变换和提升的端到端优化有效地提高了性能，同时避免了过拟合。我们还扩展我们的算法，可微的非线性变换，如果过拟合不是一个问题。为了支持可重复性和未来的工作，我们公开分享我们的代码。
摘要：We propose a soft gradient boosting framework for sequential regression that embeds a learnable linear feature transform within the boosting procedure. At each boosting iteration, we train a soft decision tree and learn a linear input feature transform Q together. This approach is particularly advantageous in high-dimensional, data-scarce scenarios, as it discovers the most relevant input representations while boosting. We demonstrate, using both synthetic and real-world datasets, that our method effectively and efficiently increases the performance by an end-to-end optimization of feature selection/transform and boosting while avoiding overfitting. We also extend our algorithm to differentiable non-linear transforms if overfitting is not a problem. To support reproducibility and future work, we share our code publicly.
【13】Towards Context-Aware Human-like Pointing Gestures with RL Motion Imitation
标题：利用RL运动模拟实现上下文感知的类人指向手势
链接：https://arxiv.org/abs/2509.12880
备注：Presented at the Context-Awareness in HRI (CONAWA) Workshop, ACM/IEEE International Conference on Human-Robot Interaction (HRI 2022), March 7, 2022
摘要：指向是与机器人交互的一种关键模式，但大多数先前的工作都集中在识别而不是生成上。我们提出了一个动作捕捉数据集的人类指向手势，涵盖不同的风格，用手习惯，和空间目标。使用带有运动模仿的强化学习，我们训练策略，在最大化精度的同时再现人类的指向。结果表明，我们的方法可以在模拟中实现上下文感知的指向行为，平衡任务性能与自然动态。
摘要：Pointing is a key mode of interaction with robots, yet most prior work has focused on recognition rather than generation. We present a motion capture dataset of human pointing gestures covering diverse styles, handedness, and spatial targets. Using reinforcement learning with motion imitation, we train policies that reproduce human-like pointing while maximizing precision. Results show our approach enables context-aware pointing behaviors in simulation, balancing task performance with natural dynamics.
【14】Gesture Evaluation in Virtual Reality
标题：虚拟现实中的手势评估
链接：https://arxiv.org/abs/2509.12816
备注：Published in Proceedings of the 26th International Conference on Multimodal Interaction (ICMI '24), ACM. Copyright 2024 ACM. Licensed under CC BY
摘要：手势是人类交流的核心，通过非语言表达丰富了互动。虚拟化身越来越多地使用人工智能生成的手势来增强逼真度，但评估主要局限于2D。虚拟现实（VR）提供了一种身临其境的替代方案，可能会影响手势的感知方式。本文对VR和2D中的计算机生成手势进行了比较评估，检查了2023年GENEA挑战赛的三个模型。结果显示，在VR中看到的手势平均评分略高，对动作捕捉“真实运动”的效果最强。“虽然模型排名在不同环境下保持一致，但VR影响了参与者的整体感知，并提供了传统2D评估的独特优势。
摘要：Gestures are central to human communication, enriching interactions through non-verbal expression. Virtual avatars increasingly use AI-generated gestures to enhance life-likeness, yet evaluations have largely been confined to 2D. Virtual Reality (VR) provides an immersive alternative that may affect how gestures are perceived. This paper presents a comparative evaluation of computer-generated gestures in VR and 2D, examining three models from the 2023 GENEA Challenge. Results show that gestures viewed in VR were rated slightly higher on average, with the strongest effect observed for motion-capture "true movement." While model rankings remained consistent across settings, VR influenced participants' overall perception and offered unique benefits over traditional 2D evaluation.
【15】Similarity-Distance-Magnitude Activations
标题：相似性-距离-幅度激活
链接：https://arxiv.org/abs/2509.12760
备注：17 pages, 5 tables, 1 algorithm. arXiv admin note: substantial text overlap with arXiv:2502.20167
摘要：We introduce a more robust and interpretable formulation of the standard softmax activation function commonly used with neural networks by adding Similarity (i.e., correctly predicted depth-matches into training) awareness and Distance-to-training-distribution awareness to the existing output Magnitude (i.e., decision-boundary) awareness. When used as the final-layer activation with language models, the resulting Similarity-Distance-Magnitude (SDM) activation function is more robust than the softmax function to co-variate shifts and out-of-distribution inputs in high-probability regions, and provides interpretability-by-exemplar via dense matching. Complementing the prediction-conditional estimates, the SDM activation enables a partitioning of the class-wise empirical CDFs to guard against low class-wise recall among selective classifications. These properties make it preferable for selective classification, even when considering post-hoc calibration methods over the softmax.
【16】Force-Modulated Visual Policy for Robot-Assisted Dressing with Arm Motions
标题：机器人辅助手臂运动穿衣的力调制视觉策略
链接：https://arxiv.org/abs/2509.12741
备注：CoRL 2025
摘要：Robot-assisted dressing has the potential to significantly improve the lives of individuals with mobility impairments. To ensure an effective and comfortable dressing experience, the robot must be able to handle challenging deformable garments, apply appropriate forces, and adapt to limb movements throughout the dressing process. Prior work often makes simplifying assumptions -- such as static human limbs during dressing -- which limits real-world applicability. In this work, we develop a robot-assisted dressing system capable of handling partial observations with visual occlusions, as well as robustly adapting to arm motions during the dressing process. Given a policy trained in simulation with partial observations, we propose a method to fine-tune it in the real world using a small amount of data and multi-modal feedback from vision and force sensing, to further improve the policy's adaptability to arm motions and enhance safety. We evaluate our method in simulation with simplified articulated human meshes and in a real world human study with 12 participants across 264 dressing trials. Our policy successfully dresses two long-sleeve everyday garments onto the participants while being adaptive to various kinds of arm motions, and greatly outperforms prior baselines in terms of task completion and user feedback. Video are available at https://dressing-motion.github.io/.
【17】MFAF: An EVA02-Based Multi-scale Frequency Attention Fusion Method for Cross-View Geo-Localization
标题：MFAF：一种基于EVA02的多尺度频率注意力融合跨视角定位方法
链接：https://arxiv.org/abs/2509.12673
备注：17 pages, 13 figures
摘要：Cross-view geo-localization aims to determine the geographical location of a query image by matching it against a gallery of images. This task is challenging due to the significant appearance variations of objects observed from variable views, along with the difficulty in extracting discriminative features. Existing approaches often rely on extracting features through feature map segmentation while neglecting spatial and semantic information. To address these issues, we propose the EVA02-based Multi-scale Frequency Attention Fusion (MFAF) method. The MFAF method consists of Multi-Frequency Branch-wise Block (MFB) and the Frequency-aware Spatial Attention (FSA) module. The MFB block effectively captures both low-frequency structural features and high-frequency edge details across multiple scales, improving the consistency and robustness of feature representations across various viewpoints. Meanwhile, the FSA module adaptively focuses on the key regions of frequency features, significantly mitigating the interference caused by background noise and viewpoint variability. Extensive experiments on widely recognized benchmarks, including University-1652, SUES-200, and Dense-UAV, demonstrate that the MFAF method achieves competitive performance in both drone localization and drone navigation tasks.
【18】Exploring Training Data Attribution under Limited Access Constraints
标题：受限访问约束下的训练数据属性研究
链接：https://arxiv.org/abs/2509.12581
摘要 ：Training data attribution (TDA) plays a critical role in understanding the influence of individual training data points on model predictions. Gradient-based TDA methods, popularized by \textit{influence function} for their superior performance, have been widely applied in data selection, data cleaning, data economics, and fact tracing. However, in real-world scenarios where commercial models are not publicly accessible and computational resources are limited, existing TDA methods are often constrained by their reliance on full model access and high computational costs. This poses significant challenges to the broader adoption of TDA in practical applications. In this work, we present a systematic study of TDA methods under various access and resource constraints. We investigate the feasibility of performing TDA under varying levels of access constraints by leveraging appropriately designed solutions such as proxy models. Besides, we demonstrate that attribution scores obtained from models without prior training on the target dataset remain informative across a range of tasks, which is useful for scenarios where computational resources are limited. Our findings provide practical guidance for deploying TDA in real-world environments, aiming to improve feasibility and efficiency under limited access.
【19】Human + AI for Accelerating Ad Localization Evaluation
标题：人类+人工智能加速广告本地化评估
链接：https://arxiv.org/abs/2509.12543
摘要：Adapting advertisements for multilingual audiences requires more than simple text translation; it demands preservation of visual consistency, spatial alignment, and stylistic integrity across diverse languages and formats. We introduce a structured framework that combines automated components with human oversight to address the complexities of advertisement localization. To the best of our knowledge, this is the first work to integrate scene text detection, inpainting, machine translation (MT), and text reimposition specifically for accelerating ad localization evaluation workflows. Qualitative results across six locales demonstrate that our approach produces semantically accurate and visually coherent localized advertisements, suitable for deployment in real-world workflows.
【20】Nonlocal Neural Tangent Kernels via Parameter-Space Interactions
标题：通过参数空间相互作用的非局部神经切核
链接：https://arxiv.org/abs/2509.12467
摘要：The Neural Tangent Kernel (NTK) framework has provided deep insights into the training dynamics of neural networks under gradient flow. However, it relies on the assumption that the network is differentiable with respect to its parameters, an assumption that breaks down when considering non-smooth target functions or parameterized models exhibiting non-differentiable behavior. In this work, we propose a Nonlocal Neural Tangent Kernel (NNTK) that replaces the local gradient with a nonlocal interaction-based approximation in parameter space. Nonlocal gradients are known to exist for a wider class of functions than the standard gradient. This allows NTK theory to be extended to nonsmooth functions, stochastic estimators, and broader families of models. We explore both fixed-kernel and attention-based formulations of this nonlocal operator. We illustrate the new formulation with numerical studies.
【21】On the Regularity and Fairness of Combinatorial Multi-Armed Bandit
标题：论组合多臂强盗的规律性与公平性
链接：https://arxiv.org/abs/2509.12457
摘要：The combinatorial multi-armed bandit model is designed to maximize cumulative rewards in the presence of uncertainty by activating a subset of arms in each round. This paper is inspired by two critical applications in wireless networks, where it's not only essential to maximize cumulative rewards but also to guarantee fairness among arms (i.e., the minimum average reward required by each arm) and ensure reward regularity (i.e., how often each arm receives the reward). In this paper, we propose a parameterized regular and fair learning algorithm to achieve these three objectives. In particular, the proposed algorithm linearly combines virtual queue-lengths (tracking the fairness violations), Time-Since-Last-Reward (TSLR) metrics, and Upper Confidence Bound (UCB) estimates in its weight measure. Here, TSLR is similar to age-of-information and measures the elapsed number of rounds since the last time an arm received a reward, capturing the reward regularity performance, and UCB estimates are utilized to balance the tradeoff between exploration and exploitation in online learning. By exploring a key relationship between virtual queue-lengths and TSLR metrics and utilizing several non-trivial Lyapunov functions, we analytically characterize zero cumulative fairness violation, reward regularity, and cumulative regret performance under our proposed algorithm. These theoretical outcomes are verified by simulations based on two real-world datasets.
【22】Evaluating the printability of stl files with ML
标题：使用ML评估stl文件的可打印性
链接：https://arxiv.org/abs/2509.12392
摘要：3D printing has long been a technology for industry professionals and enthusiasts willing to tinker or even build their own machines. This stands in stark contrast to today's market, where recent developments have prioritized ease of use to attract a broader audience. Slicing software nowadays has a few ways to sanity check the input file as well as the output gcode. Our approach introduces a novel layer of support by training an AI model to detect common issues in 3D models. The goal is to assist less experienced users by identifying features that are likely to cause print failures due to difficult to print geometries before printing even begins.
【23】Geometric Red-Teaming for Robotic Manipulation
标题：机器人操纵的几何红色团队
链接：https://arxiv.org/abs/2509.12379
备注：Accepted at the 9th Annual Conference on Robot Learning (CoRL 2025, Oral)
摘要 ：Standard evaluation protocols in robotic manipulation typically assess policy performance over curated, in-distribution test sets, offering limited insight into how systems fail under plausible variation. We introduce Geometric Red-Teaming (GRT), a red-teaming framework that probes robustness through object-centric geometric perturbations, automatically generating CrashShapes -- structurally valid, user-constrained mesh deformations that trigger catastrophic failures in pre-trained manipulation policies. The method integrates a Jacobian field-based deformation model with a gradient-free, simulator-in-the-loop optimization strategy. Across insertion, articulation, and grasping tasks, GRT consistently discovers deformations that collapse policy performance, revealing brittle failure modes missed by static benchmarks. By combining task-level policy rollouts with constraint-aware shape exploration, we aim to build a general purpose framework for structured, object-centric robustness evaluation in robotic manipulation. We additionally show that fine-tuning on individual CrashShapes, a process we refer to as blue-teaming, improves task success by up to 60 percentage points on those shapes, while preserving performance on the original object, demonstrating the utility of red-teamed geometries for targeted policy refinement. Finally, we validate both red-teaming and blue-teaming results with a real robotic arm, observing that simulated CrashShapes reduce task success from 90% to as low as 22.5%, and that blue-teaming recovers performance to up to 90% on the corresponding real-world geometry -- closely matching simulation outcomes. Videos and code can be found on our project website: https://georedteam.github.io/ .
【24】Spontaneous Kolmogorov-Arnold Geometry in Shallow MLPs
标题：浅ML中的自发Kolmogorov-Arnold几何
链接：https://arxiv.org/abs/2509.12326
备注：25 pages + 3 appendices
摘要：The Kolmogorov-Arnold (KA) representation theorem constructs universal, but highly non-smooth inner functions (the first layer map) in a single (non-linear) hidden layer neural network. Such universal functions have a distinctive local geometry, a "texture," which can be characterized by the inner function's Jacobian $J({\mathbf{x}})$, as $\mathbf{x}$ varies over the data. It is natural to ask if this distinctive KA geometry emerges through conventional neural network optimization. We find that indeed KA geometry often is produced when training vanilla single hidden layer neural networks. We quantify KA geometry through the statistical properties of the exterior powers of $J(\mathbf{x})$: number of zero rows and various observables for the minor statistics of $J(\mathbf{x})$, which measure the scale and axis alignment of $J(\mathbf{x})$. This leads to a rough understanding for where KA geometry occurs in the space of function complexity and model hyperparameters. The motivation is first to understand how neural networks organically learn to prepare input data for later downstream processing and, second, to learn enough about the emergence of KA geometry to accelerate learning through a timely intervention in network hyperparameters. This research is the "flip side" of KA-Networks (KANs). We do not engineer KA into the neural network, but rather watch KA emerge in shallow MLPs.
【25】An End to End Edge to Cloud Data and Analytics Strategy
标题：端到端边缘云数据和分析策略
链接：https://arxiv.org/abs/2509.12296
备注：None
摘要：There is an exponential growth of connected Internet of Things (IoT) devices. These have given rise to applications that rely on real time data to make critical decisions quickly. Enterprises today are adopting cloud at a rapid pace. There is a critical need to develop secure and efficient strategy and architectures to best leverage capabilities of cloud and edge assets. This paper provides an end to end secure edge to cloud data and analytics strategy. To enable real life implementation, the paper provides reference architectures for device layer, edge layer and cloud layer.
【26】RL Fine-Tuning Heals OOD Forgetting in SFT
标题：RL微调治愈SFT中的OOD遗忘
链接：https://arxiv.org/abs/2509.12235
备注：10 pages, 15 figures
摘要：The two-stage fine-tuning paradigm of Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has empirically shown better reasoning performance than one-stage SFT for the post-training of Large Language Models (LLMs). However, the evolution and mechanism behind the synergy of SFT and RL are still under-explored and inconclusive. In our study, we find the well-known claim "SFT memorizes, RL generalizes" is over-simplified, and discover that: (1) OOD performance peaks at the early stage of SFT and then declines (OOD forgetting), the best SFT checkpoint cannot be captured by training/test loss; (2) the subsequent RL stage does not generate fundamentally better OOD capability, instead it plays an \textbf{OOD restoration} role, recovering the lost reasoning ability during SFT; (3) The recovery ability has boundaries, \ie{} \textbf{if SFT trains for too short or too long, RL cannot recover the lost OOD ability;} (4) To uncover the underlying mechanisms behind the forgetting and restoration process, we employ SVD analysis on parameter matrices, manually edit them, and observe their impacts on model performance. Unlike the common belief that the shift of model capacity mainly results from the changes of singular values, we find that they are actually quite stable throughout fine-tuning. Instead, the OOD behavior strongly correlates with the \textbf{rotation of singular vectors}. Our findings re-identify the roles of SFT and RL in the two-stage fine-tuning and discover the rotation of singular vectors as the key mechanism. %reversing the rotations induced by SFT, which shows recovery from forgetting, whereas imposing the SFT parameter directions onto a RL-tuned model results in performance degradation. Code is available at https://github.com/xiaodanguoguo/RL_Heals_SFT
【27】Profiling LoRA/QLoRA Fine-Tuning Efficiency on Consumer GPUs: An RTX 4060 Case Study
标题：分析LoRA/QLoRA消费级图形处理器的微调效率：RTX 4060案例研究
链接：https://arxiv.org/abs/2509.12229
备注：8 pages, 3 figures, 2 tables. Primary category: cs.LG (Machine Learning); secondary: cs.AI (Artificial Intelligence). LaTeX source with figures included
摘要 ：Fine-tuning large language models (LLMs) with parameter-efficient techniques such as LoRA and QLoRA has enabled adaptation of foundation models on modest hardware. Yet the efficiency of such training on consumer-grade GPUs, especially under strict 8 GB VRAM limits, remains underexplored. We present a controlled profiling study of LoRA/QLoRA fine-tuning using the Qwen2.5-1.5B-Instruct model on a single NVIDIA RTX 4060. Across three representative configurations, we systematically vary batch size, sequence length, optimizer choice (AdamW vs. PagedAdamW), and precision (fp16 vs. bf16). We report throughput (tokens/s), time per 10k tokens, and VRAM footprint, alongside energy estimates derived from GPU board power limits. Our results show that paged optimizers improve throughput by up to 25% (628 tok/s vs. 500 tok/s baseline), while bf16 degrades efficiency relative to fp16. Despite 8 GB constraints, sequence lengths up to 2048 tokens were feasible using parameter-efficient strategies. To our knowledge, this is the first systematic case study of LLM fine- tuning efficiency on consumer GPUs, providing reproducible benchmarks and practical guidelines for resource-constrained researchers and practitioners.
【28】PowerGrow: Feasible Co-Growth of Structures and Dynamics for Power Grid Synthesis
标题：PowerGrow：电网综合的结构和动力学的可行协同增长
链接：https://arxiv.org/abs/2509.12212
摘要：Modern power systems are becoming increasingly dynamic, with changing topologies and time-varying loads driven by renewable energy variability, electric vehicle adoption, and active grid reconfiguration. Despite these changes, publicly available test cases remain scarce, due to security concerns and the significant effort required to anonymize real systems. Such limitations call for generative tools that can jointly synthesize grid structure and nodal dynamics. However, modeling the joint distribution of network topology, branch attributes, bus properties, and dynamic load profiles remains a major challenge, while preserving physical feasibility and avoiding prohibitive computational costs. We present PowerGrow, a co-generative framework that significantly reduces computational overhead while maintaining operational validity. The core idea is dependence decomposition: the complex joint distribution is factorized into a chain of conditional distributions over feasible grid topologies, time-series bus loads, and other system attributes, leveraging their mutual dependencies. By constraining the generation process at each stage, we implement a hierarchical graph beta-diffusion process for structural synthesis, paired with a temporal autoencoder that embeds time-series data into a compact latent space, improving both training stability and sample fidelity. Experiments across benchmark settings show that PowerGrow not only outperforms prior diffusion models in fidelity and diversity but also achieves a 98.9\% power flow convergence rate and improved N-1 contingency resilience. This demonstrates its ability to generate operationally valid and realistic power grid scenarios.
【29】QDFlow: A Python package for physics simulations of quantum dot devices
标题：QDFlow：用于量子点设备物理模拟的Python包
链接：https://arxiv.org/abs/2509.13298
备注：17 pages, 5 figures
摘要：Recent advances in machine learning (ML) have accelerated progress in calibrating and operating quantum dot (QD) devices. However, most ML approaches rely on access to large, high-quality labeled datasets for training, benchmarking, and validation, with labels capturing key features in the data. Obtaining such datasets experimentally is challenging due to limited data availability and the labor-intensive nature of labeling. QDFlow is an open-source physics simulator for multi-QD arrays that generates realistic synthetic data with ground-truth labels. QDFlow combines a self-consistent Thomas-Fermi solver, a dynamic capacitance model, and flexible noise modules to produce charge stability diagrams and ray-based data closely resembling experiments. With extensive tunable parameters and customizable noise models, QDFlow supports the creation of large, diverse datasets for ML development, benchmarking, and quantum device research.
【30】Accelerating Protein Molecular Dynamics Simulation with DeepJump
标题：使用DeepJump加速分子蛋白质动力学模拟
链接：https://arxiv.org/abs/2509.13294
摘要：Unraveling the dynamical motions of biomolecules is essential for bridging their structure and function, yet it remains a major computational challenge. Molecular dynamics (MD) simulation provides a detailed depiction of biomolecular motion, but its high-resolution temporal evolution comes at significant computational cost, limiting its applicability to timescales of biological relevance. Deep learning approaches have emerged as promising solutions to overcome these computational limitations by learning to predict long-timescale dynamics. However, generalizable kinetics models for proteins remain largely unexplored, and the fundamental limits of achievable acceleration while preserving dynamical accuracy are poorly understood. In this work, we fill this gap with DeepJump, an Euclidean-Equivariant Flow Matching-based model for predicting protein conformational dynamics across multiple temporal scales. We train DeepJump on trajectories of the diverse proteins of mdCATH, systematically studying our model's performance in generalizing to long-term dynamics of fast-folding proteins and characterizing the trade-off between computational acceleration and prediction accuracy. We demonstrate the application of DeepJump to ab initio folding, showcasing prediction of folding pathways and native states. Our results demonstrate that DeepJump achieves significant $\approx$1000$\times$ computational acceleration while effectively recovering long-timescale dynamics, providing a stepping stone for enabling routine simulation of proteins.
【31】Generalizable Holographic Reconstruction via Amplitude-Only Diffusion Priors
标题：利用纯幅度扩散先验的可推广全息重建
链接：https://arxiv.org/abs/2509.12728
备注：Keywords: Diffusion model, phase retrieval, inline-holography, inverse problem
摘要 ：Phase retrieval in inline holography is a fundamental yet ill-posed inverse problem due to the nonlinear coupling between amplitude and phase in coherent imaging. We present a novel off-the-shelf solution that leverages a diffusion model trained solely on object amplitude to recover both amplitude and phase from diffraction intensities. Using a predictor-corrector sampling framework with separate likelihood gradients for amplitude and phase, our method enables complex field reconstruction without requiring ground-truth phase data for training. We validate the proposed approach through extensive simulations and experiments, demonstrating robust generalization across diverse object shapes, imaging system configurations, and modalities, including lensless setups. Notably, a diffusion prior trained on simple amplitude data (e.g., polystyrene beads) successfully reconstructs complex biological tissue structures, highlighting the method's adaptability. This framework provides a cost-effective, generalizable solution for nonlinear inverse problems in computational imaging, and establishes a foundation for broader coherent imaging applications beyond holography.
【32】Sustainable LSTM-Based Precoding for RIS-Aided mmWave MIMO Systems with Implicit CSI
标题：具有隐式SI的RIS辅助毫米波输入输出系统的可持续基于LSTM的预编码
链接：https://arxiv.org/abs/2509.12658
备注：6 pages, 5 figures, 2 tables, and accepted by 2025 IEEE Globecom Workshops
摘要：In this paper, we propose a sustainable long short-term memory (LSTM)-based precoding framework for reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) MIMO systems. Instead of explicit channel state information (CSI) estimation, the framework exploits uplink pilot sequences to implicitly learn channel characteristics, reducing both pilot overhead and inference complexity. Practical hardware constraints are addressed by incorporating the phase-dependent amplitude model of RIS elements, while a multi-label training strategy improves robustness when multiple near-optimal codewords yield comparable performance. Simulations show that the proposed design achieves over 90% of the spectral efficiency of exhaustive search (ES) with only 2.2% of its computation time, cutting energy consumption by nearly two orders of magnitude. The method also demonstrates resilience under distribution mismatch and scalability to larger RIS arrays, making it a practical and energy-efficient solution for sustainable 6G wireless networks.
【33】Neural-Quantum-States Impurity Solver for Quantum Embedding Problems
标题：量子嵌入问题的神经量子态杂质求解器
链接：https://arxiv.org/abs/2509.12431
备注：10 pages main text, and 4 figures. Note that YinZhangHao Zhou and Zhanghao Zhouyin are the same person, I use them both
摘要：Neural quantum states (NQS) have emerged as a promising approach to solve second-quantised Hamiltonians, because of their scalability and flexibility. In this work, we design and benchmark an NQS impurity solver for the quantum embedding methods, focusing on the ghost Gutzwiller Approximation (gGA) framework. We introduce a graph transformer-based NQS framework able to represent arbitrarily connected impurity orbitals and develop an error control mechanism to stabilise iterative updates throughout the quantum embedding loops. We validate the accuracy of our approach with benchmark gGA calculations of the Anderson Lattice Model, yielding results in excellent agreement with the exact diagonalisation impurity solver. Finally, our analysis of the computational budget reveals the method's principal bottleneck to be the high-accuracy sampling of physical observables required by the embedding loop, rather than the NQS variational optimisation, directly highlighting the critical need for more efficient inference techniques.
机器翻译由腾讯交互翻译提供，仅供参考
点击“阅读原文”获取带摘要的学术速递
Python社区是高质量的Python/Django开发社区
本文地址：http://www.python88.com/topic/186859
本文地址：http://www.python88.com/topic/186859