|
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145篇
大模型相关(15篇)
【1】Asynchronous Reasoning: Training-Free Interactive Thinking LLMs
标题：同步推理：免训练互动思维法学硕士
链接：https://arxiv.org/abs/2512.10931
备注：Preprint, work in progress
摘要：许多最先进的法学硕士在给出答案之前都经过了思考。推理可以极大地提高语言模型的能力和安全性，但它也使它们的交互性降低：给定一个新的输入，模型必须停止思考才能做出响应。基于语音或嵌入式助理等真实用例需要LLM代理实时响应和适应额外的信息，这与顺序交互不兼容。相比之下，人类可以异步地倾听、思考和行动：我们在阅读问题的同时开始思考问题，在制定答案的同时继续思考。在这项工作中，我们增强了LLM的推理能力，以类似的方式操作，而无需额外的培训。我们的方法使用旋转嵌入的属性，使LLM能够同时思考，倾听和生成输出。我们评估了我们在数学、常识和安全推理方面的方法，发现它可以实时生成准确的思维增强答案，将第一个非思维令牌的时间从几分钟缩短到<= 5秒。以及6- 11倍的整体实时延迟。
摘要：Many state-of-the-art LLMs are trained to think before giving their answer. Reasoning can greatly improve language model capabilities and safety, but it also makes them less interactive: given a new input, a model must stop thinking before it can respond. Real-world use cases such as voice-based or embedded assistants require an LLM agent to respond and adapt to additional information in real time, which is incompatible with sequential interactions. In contrast, humans can listen, think, and act asynchronously: we begin thinking about the problem while reading it and continue thinking while formulating the answer. In this work, we augment LLMs capable of reasoning to operate in a similar way without additional training. Our method uses the properties of rotary embeddings to enable LLMs built for sequential interactions to simultaneously think, listen, and generate outputs. We evaluate our approach on math, commonsense, and safety reasoning and find that it can generate accurate thinking-augmented answers in real time, reducing time to first non-thinking token from minutes to <= 5s. and the overall real-time delays by 6-11x.
【2】SparseSwaps: Tractable LLM Pruning Mask Refinement at Scale
标题：SparseSwaps：可操作的LLM修剪面膜大规模优化
链接：https://arxiv.org/abs/2512.10922
备注：15 pages, 2 figures, 4 tables
摘要：神经网络的资源需求可以通过修剪来显著减少-删除看似不太重要的参数。然而，随着大型语言模型（LLM）的兴起，完全重新训练以恢复修剪引起的性能下降通常是禁止的，并且诸如全局幅度修剪的经典方法在Transformer架构上是次优的。因此，现有技术的方法解决了逐层掩模选择问题，即找到最小化校准数据的小集合上的每层修剪误差的修剪掩模的问题。由于其组合性质和搜索空间的大小，使用可编程（IP）求解器精确地将该问题求解到最优性在计算上是不可行的，因此现有方法依赖于近似或近似法。在这项工作中，我们证明了掩模选择问题可以在LLM尺度上变得更加容易处理。为此，我们通过对每行强制相等的稀疏级别来解耦行。这使我们能够导出最佳的1-交换（交换一个保留的和一个修剪的权重），可以使用校准数据的Gram矩阵有效地计算。利用这些观察结果，我们提出了一个易于处理且简单的1-swap算法，该算法从任何修剪掩码开始，在LLM规模的GPU上有效运行，并且基本上没有超参数。我们证明了我们的方法比Wanda减少了高达60%的每层修剪误差（Sun等人，2023），并不断提高困惑和zero-shot精度跨越国家的最先进的GPT架构。
摘要：The resource requirements of Neural Networks can be significantly reduced through pruning -- the removal of seemingly less important parameters. However, with the rise of Large Language Models (LLMs), full retraining to recover pruning-induced performance degradation is often prohibitive and classical approaches such as global magnitude pruning are suboptimal on Transformer architectures. State-of-the-art methods hence solve a layer-wise mask selection problem, the problem of finding a pruning mask which minimizes the per-layer pruning error on a small set of calibration data. Exactly solving this problem to optimality using Integer Programming (IP) solvers is computationally infeasible due to its combinatorial nature and the size of the search space, and existing approaches therefore rely on approximations or heuristics. In this work, we demonstrate that the mask selection problem can be made drastically more tractable at LLM scale. To that end, we decouple the rows by enforcing equal sparsity levels per row. This allows us to derive optimal 1-swaps (exchanging one kept and one pruned weight) that can be computed efficiently using the Gram matrix of the calibration data. Using these observations, we propose a tractable and simple 1-swap algorithm that warm starts from any pruning mask, runs efficiently on GPUs at LLM scale, and is essentially hyperparameter-free. We demonstrate that our approach reduces per-layer pruning error by up to 60% over Wanda (Sun et al., 2023) and consistently improves perplexity and zero-shot accuracy across state-of-the-art GPT architectures.
【3】Scaling Behavior of Discrete Diffusion Language Models
标题：离散扩散语言模型的标度行为
链接：https://arxiv.org/abs/2512.10858
摘要：现代LLM预训练消耗大量计算和训练数据，使不同模型的缩放行为或缩放定律成为关键的区分因素。离散扩散语言模型（DLM）已经被提出作为自回归语言模型（ALM）的替代。然而，它们的扩展行为尚未得到充分探索，先前的工作表明它们需要更多的数据和计算来匹配ALM的性能。 我们通过在掩蔽和均匀扩散之间平滑插值，同时密切关注关键的超参数，如批量大小和学习率，研究了DLMs在不同噪声类型上的缩放行为。我们的实验表明，DLMs的标度行为强烈依赖于噪声类型，是相当大的不同，从ALMs。虽然所有噪声类型都收敛到计算限制缩放中的类似损失值，但我们发现，与掩蔽扩散相比，均匀扩散需要更多的参数和更少的数据来进行计算有效的训练，使其成为数据限制设置中有希望的候选者。我们将我们的均匀扩散模型扩展到10 B个参数，训练了10 ^{22}$FLOP，证实了预测的缩放行为，使其成为迄今为止最大的公开已知的均匀扩散模型。
摘要：Modern LLM pre-training consumes vast amounts of compute and training data, making the scaling behavior, or scaling laws, of different models a key distinguishing factor. Discrete diffusion language models (DLMs) have been proposed as an alternative to autoregressive language models (ALMs). However, their scaling behavior has not yet been fully explored, with prior work suggesting that they require more data and compute to match the performance of ALMs. We study the scaling behavior of DLMs on different noise types by smoothly interpolating between masked and uniform diffusion while paying close attention to crucial hyperparameters such as batch size and learning rate. Our experiments reveal that the scaling behavior of DLMs strongly depends on the noise type and is considerably different from ALMs. While all noise types converge to similar loss values in compute-bound scaling, we find that uniform diffusion requires more parameters and less data for compute-efficient training compared to masked diffusion, making them a promising candidate in data-bound settings. We scale our uniform diffusion model up to 10B parameters trained for $10^{22}$ FLOPs, confirming the predicted scaling behavior and making it the largest publicly known uniform diffusion model to date.
【4】Script Gap: Evaluating LLM Triage on Indian Languages in Native vs Roman Scripts in a Real World Setting
标题：剧本差距：评估现实世界环境中原住民语言与罗马语言的LLM分类
链接：https://arxiv.org/abs/2512.10780
摘要：大型语言模型（LLM）越来越多地部署在印度的高风险临床应用中。在许多这样的环境中，印度语言的使用者经常使用罗马化文本而不是本地文字进行交流，但现有的研究很少使用真实世界的数据来评估这种拼写变化。我们调查罗马化如何影响LLM在一个关键领域的可靠性：孕产妇和新生儿医疗分流。我们基准领先的LLM上的用户生成的查询跨越五种印度语言和尼泊尔语的真实世界的数据集。我们的研究结果显示，性能的罗马化的消息，与F1分数落后于本地脚本的5-12点一致的退化。在我们在印度的合作伙伴孕产妇保健组织，这一差距可能导致近200万个额外的分诊错误。至关重要的是，脚本的这种性能差距并不是因为临床推理失败。我们证明，LLM通常正确地推断罗马化查询的语义意图。然而，在罗马化输入中存在拼写噪音的情况下，它们的最终分类输出仍然很脆弱。我们的研究结果突出了基于LLM的卫生系统中的一个关键安全盲点：似乎理解罗马化输入的模型可能仍然无法可靠地对其采取行动。
摘要：Large Language Models (LLMs) are increasingly deployed in high-stakes clinical applications in India. In many such settings, speakers of Indian languages frequently communicate using romanized text rather than native scripts, yet existing research rarely evaluates this orthographic variation using real-world data. We investigate how romanization impacts the reliability of LLMs in a critical domain: maternal and newborn healthcare triage. We benchmark leading LLMs on a real-world dataset of user-generated queries spanning five Indian languages and Nepali. Our results reveal consistent degradation in performance for romanized messages, with F1 scores trailing those of native scripts by 5-12 points. At our partner maternal health organization in India, this gap could cause nearly 2 million excess errors in triage. Crucially, this performance gap by scripts is not due to a failure in clinical reasoning. We demonstrate that LLMs often correctly infer the semantic intent of romanized queries. Nevertheless, their final classification outputs remain brittle in the presence of orthographic noise in romanized inputs. Our findings highlight a critical safety blind spot in LLM-based health systems: models that appear to understand romanized input may still fail to act on it reliably.
【5】LLM-Auction: Generative Auction towards LLM-Native Advertising
标题：法学硕士拍卖：针对法学硕士原生广告的生成性拍卖
链接：https://arxiv.org/abs/2512.10551
摘要：大型语言模型（LLM）的快速发展需要新的货币化策略，其中LLM原生广告已经成为一个有前途的范例，通过自然地将广告集成到LLM生成的响应中。然而，这种模式从根本上将拍卖对象从离散的广告位转移到LLM输出上的分布，为设计拍卖机制提出了新的挑战。现有的LLM原生广告机制采用了将拍卖和生成解耦的框架，这些框架要么忽略了外部性，要么需要多个LLM推理来进行广告分配，这使得它们在工业场景中不切实际。为了解决这些挑战，我们提出了LLM拍卖，据我们所知，这是第一个基于学习的生成拍卖机制，集成了拍卖和LLM生成LLM原生广告。通过将分配优化描述为LLM输出与反映广告商预期价值和用户体验的机制目标之间的偏好调整问题，我们引入了迭代奖励偏好优化（IRPO）算法，该算法交替优化奖励模型和LLM。这种方法使LLM能够固有地建模分配外部性，而无需任何额外的推理成本。我们进一步确定分配的单调性和连续性的LLM拍卖，这使我们能够证明，一个简单的第一价格支付规则具有良好的激励性能。此外，我们设计了一个LLM作为一个法官的模拟环境，以方便大规模的数据建设，使全面的定量评估机制的性能。大量的定量和定性实验表明，LLM-Auction显着优于现有的基线在分配效率，同时实现所需的机制属性。
摘要：The rapid advancement of large language models (LLMs) necessitates novel monetization strategies, among which LLM-native advertising has emerged as a promising paradigm by naturally integrating advertisement within LLM-generated responses. However, this paradigm fundamentally shifts the auction object from discrete ad slots to the distribution over LLM outputs, posing new challenges for designing auction mechanisms. Existing mechanisms for LLM-native advertising adopt frameworks that decouple auction and generation, which either ignore externalities or require multiple LLM inferences for ad allocation, rendering them impractical for industrial scenarios. To address these challenges, we propose LLM-Auction, which to the best of our knowledge is the first learning-based generative auction mechanism that integrates auction and LLM generation for LLM-native advertising. By formulating the allocation optimization as a preference alignment problem between LLM outputs and the mechanism's objective which reflects both advertisers' expected value and user experience, we introduce Iterative Reward-Preference Optimization (IRPO) algorithm that alternately optimizes the reward model and the LLM. This approach enables the LLM to inherently model allocation externalities without any extra inference cost. We further identify the allocation monotonicity and continuity of LLM-Auction, which allows us to prove that a simple first-price payment rule exhibits favorable incentive properties. Additionally, we design an LLM-as-a-judge simulation environment to facilitate large-scale data construction and enable comprehensive quantitative evaluation of the mechanism's performance. Extensive quantitative and qualitative experiments demonstrate that LLM-Auction significantly outperforms existing baselines in allocation efficiency, while achieving the desired mechanism properties.
【6】Unlocking the Address Book: Dissecting the Sparse Semantic Structure of LLM Key-Value Caches via Sparse Autoencoders
标题：解锁地址簿：通过稀疏自动编码器剖析LLM Key-Value缓存的稀疏语义结构
链接：https://arxiv.org/abs/2512.10547
摘要：键值（KV）缓存是长上下文大型语言模型中的主要内存瓶颈，但它通常被视为不透明的数值张量。在这项工作中，我们提出了\textbf{STA-Attention}，这是一个利用Top-K稀疏自动编码器（SAE）将KV缓存分解为可解释的语义原子的框架。与标准的$L_1$正则化SAE不同，我们的Top-K方法消除了收缩偏差，保留了注意力所需的精确点积几何。我们的分析揭示了一个基本的\textbf{Key-Value Asymmetry}：虽然Key vectors作为由"Semantic Elbow“主导的高度稀疏的路由器，但深度Value vectors携带密集的内容有效载荷，需要更大的预算。基于这种结构，我们引入了一个双预算策略，选择性地保留最翔实的语义成分，同时过滤代表性的噪音。在Yi-6 B、Mistral-7 B、Qwen 2.5 - 32 B等模型上的实验表明，我们的语义重构模型保持了与原始模型相当的困惑度和zero-shot性能，有效地弥合了机械可解释性和忠实注意建模之间的差距。
摘要：The Key-Value (KV) cache is the primary memory bottleneck in long-context Large Language Models, yet it is typically treated as an opaque numerical tensor. In this work, we propose \textbf{STA-Attention}, a framework that utilizes Top-K Sparse Autoencoders (SAEs) to decompose the KV cache into interpretable ``semantic atoms.'' Unlike standard $L_1$-regularized SAEs, our Top-K approach eliminates shrinkage bias, preserving the precise dot-product geometry required for attention. Our analysis uncovers a fundamental \textbf{Key-Value Asymmetry}: while Key vectors serve as highly sparse routers dominated by a ``Semantic Elbow,'' deep Value vectors carry dense content payloads requiring a larger budget. Based on this structure, we introduce a Dual-Budget Strategy that selectively preserves the most informative semantic components while filtering representational noise. Experiments on Yi-6B, Mistral-7B, Qwen2.5-32B, and others show that our semantic reconstructions maintain perplexity and zero-shot performance comparable to the original models, effectively bridging the gap between mechanistic interpretability and faithful attention modeling.
【7】From Lab to Reality: A Practical Evaluation of Deep Learning Models and LLMs for Vulnerability Detection
标题：从实验室到现实：深度学习模型和漏洞检测LLM的实用评估
链接：https://arxiv.org/abs/2512.10485
摘要：基于深度学习（DL）的漏洞检测方法在基准数据集上表现出了很强的性能，但它们在现实世界中的有效性仍有待探索。最近的工作表明，基于图神经网络（GNN）和基于transformer的模型，包括大型语言模型（LLM），在对策划的基准数据集进行评估时，都会产生有希望的结果。这些数据集通常以一致的数据分布和启发式或部分噪声标签为特征。在这项研究中，我们系统地评估了两个代表性的DL模型ReVeal和LineVul在四个代表性的数据集：朱丽叶，DevignBigVul和ICVul。每个模型都在每个相应的数据集上独立训练，并使用t-SNE分析它们的代码表示，以发现漏洞相关的模式。为了评估现实的适用性，我们将这些模型与四个预训练的LLM（Claude 3.5 Sonnet、GPT-o3-mini、GPT-4 o和GPT-5）一起部署在一个精心策划的数据集VentiVul上，该数据集包括20个最近（2025年5月）修复的Linux内核漏洞。我们的实验表明，目前的模型很难在表示空间中区分脆弱和非脆弱的代码，并且在具有不同分布的数据集上的泛化能力很差。当在我们新构建的时间分布外数据集VentiVul上进行评估时，性能急剧下降，大多数模型无法可靠地检测漏洞。这些结果暴露了学术基准和现实世界的部署之间的持续差距，强调了我们面向部署的评估框架的价值，以及对更强大的代码表示和更高质量的数据集的需求。
摘要：Vulnerability detection methods based on deep learning (DL) have shown strong performance on benchmark datasets, yet their real-world effectiveness remains underexplored. Recent work suggests that both graph neural network (GNN)-based and transformer-based models, including large language models (LLMs), yield promising results when evaluated on curated benchmark datasets. These datasets are typically characterized by consistent data distributions and heuristic or partially noisy labels. In this study, we systematically evaluate two representative DL models-ReVeal and LineVul-across four representative datasets: Juliet, Devign, BigVul, and ICVul. Each model is trained independently on each respective dataset, and their code representations are analyzed using t-SNE to uncover vulnerability related patterns. To assess realistic applicability, we deploy these models along with four pretrained LLMs, Claude 3.5 Sonnet, GPT-o3-mini, GPT-4o, and GPT-5 on a curated dataset, VentiVul, comprising 20 recently (May 2025) fixed vulnerabilities from the Linux kernel. Our experiments reveal that current models struggle to distinguish vulnerable from non-vulnerable code in representation space and generalize poorly across datasets with differing distributions. When evaluated on VentiVul, our newly constructed time-wise out-of-distribution dataset, performance drops sharply, with most models failing to detect vulnerabilities reliably. These results expose a persistent gap between academic benchmarks and real-world deployment, emphasizing the value of our deployment-oriented evaluation framework and the need for more robust code representations and higher-quality datasets.
【8】Dynamics of Agentic Loops in Large Language Models: A Geometric Theory of Trajectories
标题：大型语言模型中统计循环的动力学：轨迹的几何理论
链接：https://arxiv.org/abs/2512.10350
摘要：建立在大型语言模型上的递归系统通过递归反馈循环运行，其中每个输出都成为下一个输入。然而，这些代理循环的几何行为（无论它们是收敛、发散还是表现出更复杂的动力学）仍然知之甚少。本文介绍了一个分析语义嵌入空间中Agent轨迹的几何框架，将迭代变换视为离散动力系统。 我们区分的工件空间，语言转换发生，从嵌入空间，几何测量进行。由于嵌入各向异性会使余弦相似性产生偏差，因此我们引入了一种保序校准，该校准消除了系统偏差，并将相似性与人类语义判断对齐，同时保持了较高的局部稳定性。这使得轨迹，集群和吸引子的严格测量成为可能。 通过奇异agentic回路的控制实验，我们确定了两个基本制度。一个收缩重写循环收敛到一个稳定的吸引子，减少分散，而一个探索总结和否定循环产生无界发散，没有集群形成。这些制度显示出性质不同的几何签名的收缩和膨胀。 我们的研究结果表明，提示设计直接支配的动力学制度的agentic循环，使系统控制的收敛，发散和轨迹结构的迭代LLM变换。
摘要：Agentic systems built on large language models operate through recursive feedback loops, where each output becomes the next input. Yet the geometric behavior of these agentic loops (whether they converge, diverge, or exhibit more complex dynamics) remains poorly understood. This paper introduces a geometric framework for analyzing agentic trajectories in semantic embedding space, treating iterative transformations as discrete dynamical systems. We distinguish the artifact space, where linguistic transformations occur, from the embedding space, where geometric measurements are performed. Because cosine similarity is biased by embedding anisotropy, we introduce an isotonic calibration that eliminates systematic bias and aligns similarities with human semantic judgments while preserving high local stability. This enables rigorous measurement of trajectories, clusters and attractors. Through controlled experiments on singular agentic loops, we identify two fundamental regimes. A contractive rewriting loop converges toward a stable attractor with decreasing dispersion, while an exploratory summarize and negate loop produces unbounded divergence with no cluster formation. These regimes display qualitatively distinct geometric signatures of contraction and expansion. Our results show that prompt design directly governs the dynamical regime of an agentic loop, enabling systematic control of convergence, divergence and trajectory structure in iterative LLM transformations.
【9】MiniF2F-Dafny: LLM-Guided Mathematical Theorem Proving via Auto-Active Verification
标题：MiniF 2F-Dafny：通过自动主动验证进行LLM引导的数学定理证明
链接：https://arxiv.org/abs/2512.10187
摘要：我们提出了miniF 2F-Dafny，第一个翻译的数学推理基准miniF 2F自动定理证明：Dafny。以前，基准只存在于交互式定理证明器（Lean，Isabelle，HOL Light，Metamath）中。我们发现，Dafny的自动化验证了99/244（40.6%）的测试集和109/244（44.7%）的验证集与空证明-不需要手动证明步骤。对于空证明失败的问题，我们评估了12个现成的LLM提供证明提示。我们测试的最佳模型采用迭代纠错实现了55.7%的pass@4成功率。这些初步结果突出了有效的劳动分工：LLM提供高层次的指导，而自动化处理低层次的细节。我们的基准可以在GitHub上找到，网址是http://github.com/dafny-lang/miniF2F。
摘要：We present miniF2F-Dafny, the first translation of the mathematical reasoning benchmark miniF2F to an automated theorem prover: Dafny. Previously, the benchmark existed only in interactive theorem provers (Lean, Isabelle, HOL Light, Metamath). We find that Dafny's automation verifies 99/244 (40.6%) of the test set and 109/244 (44.7%) of the validation set with empty proofs--requiring no manual proof steps. For problems where empty proofs fail, we evaluate 12 off-the-shelf LLMs on providing proof hints. The best model we test achieves 55.7% pass@4 success rate employing iterative error correction. These preliminary results highlight an effective division of labor: LLMs provide high-level guidance while automation handles low-level details. Our benchmark can be found on GitHub at http://github.com/dafny-lang/miniF2F .
【10】\textsc{Text2Graph}: Combining Lightweight LLMs and GNNs for Efficient Text Classification in Label-Scarce Scenarios
标题： extSC{Text 2Shape}：结合轻量级LLM和GNN，在标签稀缺场景中实现高效文本分类
链接：https://arxiv.org/abs/2512.10061
摘要：大语言模型（LLM）已成为有效的zero-shot分类器，但其高计算要求和环境代价限制了其在高性能计算（HPC）环境中用于大规模标注的实用性。为了支持更可持续的工作流程，我们提出了\textsc{Text 2Graph}，这是一个开源的Python包，它提供了现有文本到图形分类方法的模块化实现。该框架使用户能够以灵活的方式将基于LLM的部分注释与图神经网络（GNN）标签传播相结合，从而可以直接交换组件，如特征提取器，边缘构造方法和采样策略。我们使用跨越主题分类和情感分析任务的五个数据集在zero-shot设置上对\textsc{Text 2Graph}进行基准测试，将多个变体与其他文本分类的zero-shot方法进行比较。除了报告性能外，我们还提供了能源消耗和碳排放的详细估计，表明基于图的传播以能源和环境成本的一小部分实现了有竞争力的结果。
摘要 ：Large Language Models (LLMs) have become effective zero-shot classifiers, but their high computational requirements and environmental costs limit their practicality for large-scale annotation in high-performance computing (HPC) environments. To support more sustainable workflows, we present \textsc{Text2Graph}, an open-source Python package that provides a modular implementation of existing text-to-graph classification approaches. The framework enables users to combine LLM-based partial annotation with Graph Neural Network (GNN) label propagation in a flexible manner, making it straightforward to swap components such as feature extractors, edge construction methods, and sampling strategies. We benchmark \textsc{Text2Graph} on a zero-shot setting using five datasets spanning topic classification and sentiment analysis tasks, comparing multiple variants against other zero-shot approaches for text classification. In addition to reporting performance, we provide detailed estimates of energy consumption and carbon emissions, showing that graph-based propagation achieves competitive results at a fraction of the energy and environmental cost.
【11】Detailed balance in large language model-driven agents
标题：大型语言模型驱动代理中的详细平衡
链接：https://arxiv.org/abs/2512.10047
备注：20 pages, 12 figures, 5 tables
摘要：大型语言模型（LLM）驱动的代理正在成为解决复杂问题的一个强大的新范式。尽管这些实践在经验上取得了成功，但仍然缺乏理解和统一其宏观动态的理论框架。这封信提出了一种基于最小作用原理的方法来估计嵌入在代理中的LLM的潜在生成方向性。通过实验测量LLM生成的状态之间的转换概率，我们在统计上发现了LLM生成的转换中的详细平衡，表明LLM生成可能不是通过一般学习规则集和策略来实现的，而是通过隐式学习一类可能超越不同LLM架构和提示模板的潜在函数来实现的。据我们所知，这是在LLM生成动力学中首次发现不依赖于特定模型细节的宏观物理定律。这项工作试图建立复杂人工智能系统的宏观动力学理论，旨在将人工智能主体的研究从工程实践的集合提升到建立在可预测和可量化的有效测量基础上的科学。
摘要：Large language model (LLM)-driven agents are emerging as a powerful new paradigm for solving complex problems. Despite the empirical success of these practices, a theoretical framework to understand and unify their macroscopic dynamics remains lacking. This Letter proposes a method based on the least action principle to estimate the underlying generative directionality of LLMs embedded within agents. By experimentally measuring the transition probabilities between LLM-generated states, we statistically discover a detailed balance in LLM-generated transitions, indicating that LLM generation may not be achieved by generally learning rule sets and strategies, but rather by implicitly learning a class of underlying potential functions that may transcend different LLM architectures and prompt templates. To our knowledge, this is the first discovery of a macroscopic physical law in LLM generative dynamics that does not depend on specific model details. This work is an attempt to establish a macroscopic dynamics theory of complex AI systems, aiming to elevate the study of AI agents from a collection of engineering practices to a science built on effective measurements that are predictable and quantifiable.
【12】Local LLM Ensembles for Zero-shot Portuguese Named Entity Recognition
标题：Zero-Shot葡萄牙命名实体识别本地LLM合奏
链接：https://arxiv.org/abs/2512.10043
摘要：大型语言模型（LLM）通过上下文学习在许多自然语言处理（NLP）任务中表现出色，但在命名实体识别（NER）中表现不佳，特别是对于葡萄牙语等资源较低的语言。虽然开放权重LLM支持本地部署，但没有一个模型可以主导所有任务，这激发了集成方法。然而，现有的LLM合奏集中在文本生成或分类，留下NER探索不足。在这种情况下，这项工作提出了一种新的三步合奏管道zero-shot NER使用类似的能力，本地运行的LLM。我们的方法优于个人LLM在四个葡萄牙NER数据集利用启发式选择最佳模型组合与最小的注释数据。此外，我们表明，在不同的源数据集上获得的集合通常优于跨数据集配置中的单个LLM，从而可能消除当前任务对注释数据的需求。我们的工作进展可扩展的，低资源，和zero-shot NER有效地结合多个小LLM没有微调。代码可在https://github.com/Joao-Luz/local-llm-ner-ensemble上获得。
摘要：Large Language Models (LLMs) excel in many Natural Language Processing (NLP) tasks through in-context learning but often under-perform in Named Entity Recognition (NER), especially for lower-resource languages like Portuguese. While open-weight LLMs enable local deployment, no single model dominates all tasks, motivating ensemble approaches. However, existing LLM ensembles focus on text generation or classification, leaving NER under-explored. In this context, this work proposes a novel three-step ensemble pipeline for zero-shot NER using similarly capable, locally run LLMs. Our method outperforms individual LLMs in four out of five Portuguese NER datasets by leveraging a heuristic to select optimal model combinations with minimal annotated data. Moreover, we show that ensembles obtained on different source datasets generally outperform individual LLMs in cross-dataset configurations, potentially eliminating the need for annotated data for the current task. Our work advances scalable, low-resource, and zero-shot NER by effectively combining multiple small LLMs without fine-tuning. Code is available at https://github.com/Joao-Luz/local-llm-ner-ensemble.
【13】Intelligently Weighting Multiple Reference Models for Direct Preference Optimization of LLMs
标题：智能加权多个参考模型以实现LLM的直接偏好优化
链接：https://arxiv.org/abs/2512.10040
备注：Working paper. 13 pages, 4 figures
摘要：微调是将大型语言模型（LLM）与人类偏好对齐的必要条件。多参考偏好优化（MRPO）建立在直接偏好优化（DPO）的基础上，通过微调偏好数据集上的LLM，同时将策略正则化为参考模型的混合，以利用它们的集体期望属性。然而，目前用于设置参考权重的方法是临时的，并且在统计上是不合理的，导致不可靠的性能。为了解决这个问题，我们引入了四个新的加权策略：两个离线的方法，利用保持了验证信号;一个在线的方法，使用滑动窗口估计，以减少过拟合;和一个在线的方法，将参考加权作为一个$K$-武装土匪通过汤普森采样。使用Qwen2.5-0.5B作为策略模型和来自Llama，Mistral，Qwen，Yi和Phi家族的7个参考模型（每个0.5B-14 B）的实验表明，我们的4种策略在偏好准确性方面都优于UltraFeedback和SafeRLHF上当前的MRPO加权方法。然而，更发人深省的是，我们发现单引用DPO，使用7个引用中的6个，始终优于所有测试的多引用方法-质疑多引用方法的实际吸引力。
摘要：Fine-tuning is integral for aligning large language models (LLMs) with human preferences. Multiple-Reference Preference Optimization (MRPO) builds on Direct Preference Optimization (DPO) by fine-tuning LLMs on preference datasets while regularizing the policy towards a mixture of reference models to leverage their collective desirable properties. However, current methods for setting the reference weights are ad-hoc and statistically unsound, leading to unreliable performance. To address this, we introduce four new weighting strategies: two offline methods that leverage held-out validation signal; one online method that uses a sliding-window estimator to reduce overfitting; and an online method that treats reference weighting as a $K$-armed bandit via Thompson Sampling. Experiments using Qwen2.5-0.5B as the policy model and seven reference models from the Llama, Mistral, Qwen, Yi, and Phi families (0.5B-14B each) show that all 4 of our strategies outperform the current MRPO weighting methods on UltraFeedback and SafeRLHF in preference accuracy. More thought-provokingly, however, we find that single-reference DPO, using any of 6 out of 7 references, consistently outperforms all tested multiple-reference approaches -- calling into question the practical appeal of multiple-reference approaches.
【14】BAMBO: Construct Ability and Efficiency LLM Pareto Set via Bayesian Adaptive Multi-objective Block-wise Optimization
标题：BAMBO：通过Bayesian自适应多目标块优化构建能力和效率LLM Pareto集
链接：https://arxiv.org/abs/2512.09972
摘要 ：构建一个帕累托集是导航大型语言模型（LLM）的能力-效率权衡的关键;然而，现有的合并技术仍然不足以完成这项任务。粗粒度的模型级方法只产生稀疏的次优解集，而细粒度的分层方法则受到“维数灾难”的影响，使得搜索空间在计算上难以处理。为了解决这个二分法，我们提出了BAMBO（贝叶斯自适应多目标块优化），一个新的框架，自动构建LLM帕累托集。BAMBO通过引入混合最优块划分策略使搜索易于处理。作为一个一维聚类问题，该策略利用动态规划方法来最佳地平衡块内同质性和块间信息分布，从而在不牺牲关键粒度的情况下显着降低维数。整个过程在由q-预期超容积改善（qEHVI）采集功能驱动的进化循环内自动进行。实验表明，BAMBO发现了一个优于基线和更全面的帕累托边界，使敏捷的模型选择量身定制，以适应不同的运营约束。代码可从以下网址获得：https://github.com/xin8coder/BAMBO.
摘要：Constructing a Pareto set is pivotal for navigating the capability-efficiency trade-offs in Large Language Models (LLMs); however, existing merging techniques remain inadequate for this task. Coarse-grained, model-level methods yield only a sparse set of suboptimal solutions, while fine-grained, layer-wise approaches suffer from the "curse of dimensionality," rendering the search space computationally intractable. To resolve this dichotomy, we propose BAMBO (Bayesian Adaptive Multi-objective Block-wise Optimization), a novel framework that automatically constructs the LLM Pareto set. BAMBO renders the search tractable by introducing a Hybrid Optimal Block Partitioning strategy. Formulated as a 1D clustering problem, this strategy leverages a dynamic programming approach to optimally balance intra-block homogeneity and inter-block information distribution, thereby dramatically reducing dimensionality without sacrificing critical granularity. The entire process is automated within an evolutionary loop driven by the q-Expected Hypervolume Improvement (qEHVI) acquisition function. Experiments demonstrate that BAMBO discovers a superior and more comprehensive Pareto frontier than baselines, enabling agile model selection tailored to diverse operational constraints. Code is available at: https://github.com/xin8coder/BAMBO.
【15】Planning, Living and Judging: A Multi-agent LLM-based Framework for Cyclical Urban Planning
标题：规划、生活和判断：基于多主体法学硕士的周期性城市规划框架
链接：https://arxiv.org/abs/2412.20505
备注：4 pages, 2 figures, accepted by The 1st Workshop on AI for Urban Planning (AAAI 2025's Workshop)
摘要：城市复兴在城市化背景下提出了重大挑战，需要采取适应性办法来应对不断变化的需求。利用大型语言模型（LLM）的进步，我们提出了循环城市规划（CUP），这是一种新的范式，可以在闭环中不断生成，评估和完善城市规划。具体来说，我们基于多智能体的LLM框架由三个关键组成部分组成：（1）规划，LLM代理根据上下文数据生成和完善城市规划;（2）生活，代理模拟居民的行为和互动，模拟城市环境中的生活;（3）判断，包括评估计划有效性并提供迭代反馈以进行改进。周期性的过程使规划方法具有动态性和响应性。在真实世界数据集上的实验证明了我们的框架作为一个连续和自适应的规划过程的有效性。
摘要：Urban regeneration presents significant challenges within the context of urbanization, requiring adaptive approaches to tackle evolving needs. Leveraging advancements in large language models (LLMs), we propose Cyclical Urban Planning (CUP), a new paradigm that continuously generates, evaluates, and refines urban plans in a closed-loop. Specifically, our multi-agent LLM-based framework consists of three key components: (1) Planning, where LLM agents generate and refine urban plans based on contextual data; (2) Living, where agents simulate the behaviors and interactions of residents, modeling life in the urban environment; and (3) Judging, which involves evaluating plan effectiveness and providing iterative feedback for improvement. The cyclical process enables a dynamic and responsive planning approach. Experiments on the real-world dataset demonstrate the effectiveness of our framework as a continuous and adaptive planning process.
Graph相关(图学习|图神经网络|图优化等)(7篇)
【1】Template-Free Retrosynthesis with Graph-Prior Augmented Transformers
标题：使用图形优先增强Transformer的无模板逆合成
链接：https://arxiv.org/abs/2512.10770
摘要：逆合成反应预测是计算机辅助有机合成中的一个核心问题，它试图推断出给定产物的可能反应物分子。尽管最近取得了进展，但许多现有模型仍然达不到实际部署所需的准确性和鲁棒性。这项工作研究了一个无模板，基于transformer的框架，消除了对手工制作的反应模板或额外的化学规则引擎的依赖。该模型将分子图信息注入注意力机制，联合利用SMILES序列和结构线索，并进一步应用配对数据增强策略，提高训练的多样性和规模。在USPTO-50 K基准测试中，我们提出的方法在无模板方法中实现了最先进的性能，并且大大优于普通Transformer基线。
摘要：Retrosynthesis reaction prediction seeks to infer plausible reactant molecules for a given product and is a central problem in computer-aided organic synthesis. Despite recent progress, many existing models still fall short of the accuracy and robustness required for practical deployment. This work studies a template-free, Transformer-based framework that eliminates reliance on handcrafted reaction templates or additional chemical rule engines. The model injects molecular graph information into the attention mechanism to jointly exploit \SMILES\ sequences and structural cues, and further applies a paired data augmentation strategy to enhance training diversity and scale. On the USPTO-50K benchmark, our proposed approach achieves state-of-the-art performance among template-free methods and substantially outperforming a vanilla Transformer baseline.
【2】LGAN: An Efficient High-Order Graph Neural Network via the Line Graph Aggregation
标题：LGAN：一种通过线性图聚合的高效高级图神经网络
链接：https://arxiv.org/abs/2512.10735
摘要：图神经网络（GNN）已经成为图分类的主要范式。具体来说，大多数现有的GNN主要依赖于相邻节点之间的消息传递策略，其中表达性受到一维Weisfeiler-Lehman（1-WL）测试的限制。尽管已经提出了许多基于k-WL的GNN来克服这一限制，但它们的计算成本随着k的增加而迅速增加，这大大限制了实际适用性。此外，由于k-WL模型主要对节点元组进行操作，因此这些基于k-WL的GNN无法保留属性方法所需的细粒度节点或边级语义（例如，综合因素），导致解释性较低的问题。为了克服上述缺点，本文提出了一种新的线图聚合网络（LGAN），该网络从以每个节点为中心的导出子图构造线图，进行高阶聚合。我们从理论上证明了LGAN不仅在内射聚集假设下具有比2-WL更强的表达能力，而且具有更低的时间复杂度。对基准的实证评估表明，LGAN优于最先进的基于k-WL的GNN，同时提供更好的可解释性。
摘要：Graph Neural Networks (GNNs) have emerged as a dominant paradigm for graph classification. Specifically, most existing GNNs mainly rely on the message passing strategy between neighbor nodes, where the expressivity is limited by the 1-dimensional Weisfeiler-Lehman (1-WL) test. Although a number of k-WL-based GNNs have been proposed to overcome this limitation, their computational cost increases rapidly with k, significantly restricting the practical applicability. Moreover, since the k-WL models mainly operate on node tuples, these k-WL-based GNNs cannot retain fine-grained node- or edge-level semantics required by attribution methods (e.g., Integrated Gradients), leading to the less interpretable problem. To overcome the above shortcomings, in this paper, we propose a novel Line Graph Aggregation Network (LGAN), that constructs a line graph from the induced subgraph centered at each node to perform the higher-order aggregation. We theoretically prove that the LGAN not only possesses the greater expressive power than the 2-WL under injective aggregation assumptions, but also has lower time complexity. Empirical evaluations on benchmarks demonstrate that the LGAN outperforms state-of-the-art k-WL-based GNNs, while offering better interpretability.
【3】THeGAU: Type-Aware Heterogeneous Graph Autoencoder and Augmentation
标题：TheGAU：类型感知的异类图形自动编码器和增强
链接：https://arxiv.org/abs/2512.10589
摘要：异构图神经网络（HGNNs）是对异构信息网络（HIN）进行建模的有效方法，HIN编码复杂的多类型实体和关系。然而，HGNN经常遭受类型信息丢失和结构噪声，限制了它们的表示保真度和泛化能力。我们提出了THeGAU，一个模型无关的框架，它结合了类型感知的图形自动编码器和引导的图形增强来提高节点分类。THeGAU重建模式有效的边缘作为一个辅助任务，以保持节点类型的语义，并引入了一个解码器驱动的增强机制，有选择地细化噪声结构。这种联合设计增强了鲁棒性、准确性和效率，同时显著降低了计算开销。在三个基准HIN数据集（IMDB，ACM和DBLP）上进行的大量实验表明，THeGAU始终优于现有的HGNN方法，在多个主干上实现了最先进的性能。
摘要：Heterogeneous Graph Neural Networks (HGNNs) are effective for modeling Heterogeneous Information Networks (HINs), which encode complex multi-typed entities and relations. However, HGNNs often suffer from type information loss and structural noise, limiting their representational fidelity and generalization. We propose THeGAU, a model-agnostic framework that combines a type-aware graph autoencoder with guided graph augmentation to improve node classification. THeGAU reconstructs schema-valid edges as an auxiliary task to preserve node-type semantics and introduces a decoder-driven augmentation mechanism to selectively refine noisy structures. This joint design enhances robustness, accuracy, and efficiency while significantly reducing computational overhead. Extensive experiments on three benchmark HIN datasets (IMDB, ACM, and DBLP) demonstrate that THeGAU consistently outperforms existing HGNN methods, achieving state-of-the-art performance across multiple backbones.
【4】Better Prevent than Tackle: Valuing Defense in Soccer Based on Graph Neural Networks
标题：防胜于攻：基于图神经网络的足球防守价值评估
链接：https://arxiv.org/abs/2512.10355
摘要：评估足球的防守表现仍然具有挑战性，因为有效的防守通常不是通过拦截和铲球等可见的球上行动来表达的，而是通过在危险机会出现之前阻止它们。现有的方法在很大程度上集中在评估球上的行动，留下了很多后卫的真正影响无法衡量。为了解决这一差距，我们提出了DEFCON（防守CONSIDENCE评估器），一个全面的框架，量化球员级别的防守贡献，在足球的每一个攻击的情况。DEFCON利用图形注意力网络（Graph Attention Networks）来评估每个攻击选项的成功概率和期望值，以及每个防守者阻止它的责任。这些组件在每个行动之前和之后为攻击队产生预期占有值（EPV），DEFCON根据防守者是否减少或增加对手的EPV来分配正或负的学分。在2023-24赛季进行训练，并根据2024-25赛季的埃雷迪维西赛事和跟踪数据进行评估，DEFCON的球员积分总和与市场估值呈现出强烈的正相关性。最后，我们展示了几个实际的应用，包括在游戏中的时间表的防守贡献，跨球场区域的空间分析，和成对的攻击者-防守者的互动总结。
摘要：Evaluating defensive performance in soccer remains challenging, as effective defending is often expressed not through visible on-ball actions such as interceptions and tackles, but through preventing dangerous opportunities before they arise. Existing approaches have largely focused on valuing on-ball actions, leaving much of defenders' true impact unmeasured. To address this gap, we propose DEFCON (DEFensive CONtribution evaluator), a comprehensive framework that quantifies player-level defensive contributions for every attacking situation in soccer. Leveraging Graph Attention Networks, DEFCON estimates the success probability and expected value of each attacking option, along with each defender's responsibility for stopping it. These components yield an Expected Possession Value (EPV) for the attacking team before and after each action, and DEFCON assigns positive or negative credits to defenders according to whether they reduced or increased the opponent's EPV. Trained on 2023-24 and evaluated on 2024-25 Eredivisie event and tracking data, DEFCON's aggregated player credits exhibit strong positive correlations with market valuations. Finally, we showcase several practical applications, including in-game timelines of defensive contributions, spatial analyses across pitch zones, and pairwise summaries of attacker-defender interactions.
【5】HGC-Herd: Efficient Heterogeneous Graph Condensation via Representative Node Herding
标题：HGC-Herd：通过代表性节点羊群进行高效的异类图凝聚
链接：https://arxiv.org/abs/2512.09947
备注：8 pages, 2 figures
摘要：异构图神经网络（HGNNs）在跨多类型节点和关系的复杂语义建模方面表现出强大的能力。然而，由于结构冗余和高维节点特征，它们对大规模图的可扩展性仍然具有挑战性。现有的图压缩方法，如GCond，主要是针对同构图开发的，并且依赖于梯度匹配，导致相当大的计算，内存和优化开销。我们提出了HGC-Herd，一个无需训练的压缩框架，它可以生成紧凑而信息丰富的异构图，同时保持语义和结构的保真度。HGC-Herd集成了轻量级特征传播来编码多跳关系上下文，并采用类式放牧机制来识别每个类的代表节点，为下游学习任务产生平衡和有区别的子集。在ACM、DBLP和Freebase上进行的大量实验验证了HGC-Herd达到了与全图训练相当或更高的精度，同时显著降低了运行时间和内存消耗。这些结果强调了它的实用价值，高效和可扩展的异构图表示学习。
摘要：Heterogeneous graph neural networks (HGNNs) have demonstrated strong capability in modeling complex semantics across multi-type nodes and relations. However, their scalability to large-scale graphs remains challenging due to structural redundancy and high-dimensional node features. Existing graph condensation approaches, such as GCond, are primarily developed for homogeneous graphs and rely on gradient matching, resulting in considerable computational, memory, and optimization overhead. We propose HGC-Herd, a training-free condensation framework that generates compact yet informative heterogeneous graphs while maintaining both semantic and structural fidelity. HGC-Herd integrates lightweight feature propagation to encode multi-hop relational context and employs a class-wise herding mechanism to identify representative nodes per class, producing balanced and discriminative subsets for downstream learning tasks. Extensive experiments on ACM, DBLP, and Freebase validate that HGC-Herd attains comparable or superior accuracy to full-graph training while markedly reducing both runtime and memory consumption. These results underscore its practical value for efficient and scalable heterogeneous graph representation learning.
【6】CC-GRMAS: A Multi-Agent Graph Neural System for Spatiotemporal Landslide Risk Assessment in High Mountain Asia
标题：CC-GRMAS：亚洲高山时空滑坡风险评估的多智能体图神经系统
链接：https://arxiv.org/abs/2510.20875
摘要：滑坡是一种日益严重的气候诱发灾害，对环境和人类造成严重后果，特别是在亚洲高山地区。尽管越来越多地获得卫星和时间数据集，及时检测和灾害应对仍然不够发达和分散。这项工作介绍了CC-GRMAS，一个框架，利用一系列的卫星观测和环境信号，以提高滑坡预测的准确性。该系统围绕三个相互关联的代理预测，规划和执行，协同实现实时态势感知，响应规划和干预。通过纳入当地环境因素并实施多代理协调，这种方法为脆弱山区的气候适应性灾害准备提供了一个可扩展和积极主动的解决方案。
摘要：Landslides are a growing climate induced hazard with severe environmental and human consequences, particularly in high mountain Asia. Despite increasing access to satellite and temporal datasets, timely detection and disaster response remain underdeveloped and fragmented. This work introduces CC-GRMAS, a framework leveraging a series of satellite observations and environmental signals to enhance the accuracy of landslide forecasting. The system is structured around three interlinked agents Prediction, Planning, and Execution, which collaboratively enable real time situational awareness, response planning, and intervention. By incorporating local environmental factors and operationalizing multi agent coordination, this approach offers a scalable and proactive solution for climate resilient disaster preparedness across vulnerable mountainous terrains.
【7】Topology-Guided Quantum GANs for Constrained Graph Generation
标题：用于约束图生成的用于映射的量子GAN
链接：https://arxiv.org/abs/2512.10582
摘要：量子计算（QC）具有理论上的优势，有利于那些无法有效地经典模拟的计算问题。然而，这种理论上的加速在很大程度上取决于解决问题的量子电路设计。我们认为，质量控制文献尚未探索更多的领域特定的ansatz拓扑结构，而不是依赖于通用的，一刀切的架构。在这项工作中，我们表明，将特定任务的归纳偏差-特别是几何先验-纳入量子电路设计中，可以增强混合量子生成对抗网络（QuGAN）在生成几何约束K4图的任务上的性能。我们评估了一个组合的纠缠拓扑结构和损失函数的设计，以评估其对统计保真度和遵守几何约束，包括三角形和托勒密不等式的影响。我们的研究结果表明，将电路拓扑结构与底层问题结构对齐会产生实质性的好处：三角拓扑Qu GAN在量子模型中实现了最高的几何有效性，并与经典生成对抗网络（GAN）的性能相匹配。此外，我们展示了特定的架构选择，如纠缠门类型，方差正则化和输出缩放管理几何一致性和分布精度之间的权衡，从而强调结构化，任务感知的量子ansatz拓扑的价值。
摘要：Quantum computing (QC) promises theoretical advantages, benefiting computational problems that would not be efficiently classically simulatable. However, much of this theoretical speedup depends on the quantum circuit design solving the problem. We argue that QC literature has yet to explore more domain specific ansatz-topologies, instead of relying on generic, one-size-fits-all architectures. In this work, we show that incorporating task-specific inductive biases -- specifically geometric priors -- into quantum circuit design can enhance the performance of hybrid Quantum Generative Adversarial Networks (QuGANs) on the task of generating geometrically constrained K4 graphs. We evaluate a portfolio of entanglement topologies and loss-function designs to assess their impact on both statistical fidelity and compliance with geometric constraints, including the Triangle and Ptolemaic inequalities. Our results show that aligning circuit topology with the underlying problem structure yields substantial benefits: the Triangle-topology QuGAN achieves the highest geometric validity among quantum models and matches the performance of classical Generative Adversarial Networks (GAN). Additionally, we showcase how specific architectural choices, such as entangling gate types, variance regularization and output-scaling govern the trade-off between geometric consistency and distributional accuracy, thus emphasizing the value of structured, task-aware quantum ansatz-topologies.
Transformer(4篇)
【1】Stronger Normalization-Free Transformers
标题：更强的无规范化Transformer
链接：https://arxiv.org/abs/2512.10938
摘要：尽管规范化层长期以来一直被视为深度学习架构不可或缺的组件，但最近引入的动态Tanh（DyT）已经证明了替代方案是可能的。逐点函数DyT约束极值以实现稳定收敛并达到归一化水平的性能;这项工作进一步寻求可以超越它的函数设计。我们首先研究了逐点函数的内在属性如何影响训练和性能。基于这些发现，我们进行了一个更有效的功能设计大规模的搜索。通过这种探索，我们引入了$\mathrm{Derf}（x）= \mathrm{erf}（αx + s）$，其中$\mathrm{erf}（x）$是重新缩放的高斯累积分布函数，并将其确定为最佳性能设计。Derf在许多领域都优于LayerNorm、RMSNorm和DyT，包括视觉（图像识别和生成）、语音表示和DNA序列建模。我们的研究结果表明，Derf的性能增益主要源于其改进的泛化能力，而不是更强的拟合能力。它的简单性和更强的性能使Derf成为无规范化Transformer架构的实用选择。
摘要：Although normalization layers have long been viewed as indispensable components of deep learning architectures, the recent introduction of Dynamic Tanh (DyT) has demonstrated that alternatives are possible. The point-wise function DyT constrains extreme values for stable convergence and reaches normalization-level performance; this work seeks further for function designs that can surpass it. We first study how the intrinsic properties of point-wise functions influence training and performance. Building on these findings, we conduct a large-scale search for a more effective function design. Through this exploration, we introduce $\mathrm{Derf}(x) = \mathrm{erf}(αx + s)$, where $\mathrm{erf}(x)$ is the rescaled Gaussian cumulative distribution function, and identify it as the most performant design. Derf outperforms LayerNorm, RMSNorm, and DyT across a wide range of domains, including vision (image recognition and generation), speech representation, and DNA sequence modeling. Our findings suggest that the performance gains of Derf largely stem from its improved generalization rather than stronger fitting capacity. Its simplicity and stronger performance make Derf a practical choice for normalization-free Transformer architectures.
【2】UrbanAI 2025 Challenge: Linear vs Transformer Models for Long-Horizon Exogenous Temperature Forecasting
标题：UrbanAI 2025挑战：线性模型与Transformer模型用于长期外生温度预测
链接：https://arxiv.org/abs/2512.10866
备注：NeurIPS 2025 Workshop UrbanAI
摘要：我们研究了长期的外生温度预测-一个具有挑战性的单变量设置，其中只有室内温度的过去值用于预测-使用线性和变压器系列模型。我们根据标准化的训练、验证和测试分割来评估Linear、NLinear、DLinear、Transformer、Informer和Autoformer。结果表明，线性基线（线性，NLinear，DLinear）始终优于更复杂的Transformer系列架构，DLinear在所有分割中实现了最佳的整体准确性。这些研究结果强调，精心设计的线性模型仍然是具有挑战性的外部环境中时间序列预测的强基线。
摘要：We study long-horizon exogenous-only temperature forecasting - a challenging univariate setting where only the past values of the indoor temperature are used for prediction - using linear and Transformer-family models. We evaluate Linear, NLinear, DLinear, Transformer, Informer, and Autoformer under standardized train, validation, and test splits. Results show that linear baselines (Linear, NLinear, DLinear) consistently outperform more complex Transformer-family architectures, with DLinear achieving the best overall accuracy across all splits. These findings highlight that carefully designed linear models remain strong baselines for time series forecasting in challenging exogenous-only settings.
【3】GPG: Generalized Policy Gradient Theorem for Transformer-based Policies
标题：GPG：基于转换器的政策的广义政策梯度定理
链接：https://arxiv.org/abs/2512.10365
摘要：我们提出了广义策略梯度（GPG）定理，专门为基于转换器的策略设计。值得注意的是，我们证明了标准的政策梯度定理和GRPO出现在我们的GPG框架内的特殊情况。此外，我们还探索了它在训练大型语言模型（LLM）方面的实际应用，为有效的策略优化提供了新的见解。
摘要：We present the Generalized Policy Gradient (GPG) Theorem, specifically designed for Transformer-based policies. Notably, we demonstrate that both standard Policy Gradient Theorem and GRPO emerge as special cases within our GPG framework. Furthermore, we explore its practical applications in training Large Language Models (LLMs), offering new insights into efficient policy optimization.
【4】QSTAformer: A Quantum-Enhanced Transformer for Robust Short-Term Voltage Stability Assessment against Adversarial Attacks
标题
：QSTAformer：一种量子增强型Transformer，用于针对对抗攻击进行稳健的短期电压稳定性评估
链接：https://arxiv.org/abs/2512.09936
备注：15 pages, 12 figures. Accepted by Applied Energy
摘要：短期电压稳定评估（STVSA）是保证电力系统安全运行的重要手段。虽然经典的基于机器学习的方法已经表现出强大的性能，但它们在对抗性条件下仍然面临鲁棒性的挑战。本文提出了QSTAformer-一个定制的量子增强的Transformer架构，嵌入参数化的量子电路（PQC）到注意力机制-强大而有效的STVSA。一个专门的对抗性训练策略，以抵御白盒和灰盒攻击。此外，不同的PQC体系结构进行基准测试，以探索表现力，收敛性和效率之间的权衡。据我们所知，这是第一次系统地研究基于量子机器学习的STVSA的对抗脆弱性。对IEEE 39节点系统的案例研究表明，QSTAformer实现了具有竞争力的准确性，降低了复杂性和更强的鲁棒性，强调了其在对抗条件下安全和可扩展的STVSA的潜力。
摘要：Short-term voltage stability assessment (STVSA) is critical for secure power system operation. While classical machine learning-based methods have demonstrated strong performance, they still face challenges in robustness under adversarial conditions. This paper proposes QSTAformer-a tailored quantum-enhanced Transformer architecture that embeds parameterized quantum circuits (PQCs) into attention mechanisms-for robust and efficient STVSA. A dedicated adversarial training strategy is developed to defend against both white-box and gray-box attacks. Furthermore, diverse PQC architectures are benchmarked to explore trade-offs between expressiveness, convergence, and efficiency. To the best of our knowledge, this is the first work to systematically investigate the adversarial vulnerability of quantum machine learning-based STVSA. Case studies on the IEEE 39-bus system demonstrate that QSTAformer achieves competitive accuracy, reduced complexity, and stronger robustness, underscoring its potential for secure and scalable STVSA under adversarial conditions.
GAN|对抗|攻击|生成相关(8篇)
【1】Empirical evaluation of the Frank-Wolfe methods for constructing white-box adversarial attacks
标题：构建白盒对抗攻击的Frank-Wolfe方法的经验评估
链接：https://arxiv.org/abs/2512.10936
摘要：神经网络对抗性攻击的构建似乎是其在各种服务中部署的关键挑战。为了评估神经网络的对抗鲁棒性，需要一种快速有效的方法来构造对抗攻击。由于对抗性攻击构造的形式化涉及解决特定的优化问题，因此我们从数值优化的角度考虑构造高效和有效的对抗性攻击的问题。具体来说，我们建议利用先进的无投影方法，即改进的Frank-Wolfe方法，对给定的输入数据进行白盒对抗攻击。我们对这些方法进行了理论和数值评估，并将它们与基于投影操作或几何直观的标准方法进行了比较。在MNIST和CIFAR-10数据集上进行数值实验，利用多类逻辑回归模型，卷积神经网络（CNN）和Vision Transformer（ViT）。
摘要：The construction of adversarial attacks for neural networks appears to be a crucial challenge for their deployment in various services. To estimate the adversarial robustness of a neural network, a fast and efficient approach is needed to construct adversarial attacks. Since the formalization of adversarial attack construction involves solving a specific optimization problem, we consider the problem of constructing an efficient and effective adversarial attack from a numerical optimization perspective. Specifically, we suggest utilizing advanced projection-free methods, known as modified Frank-Wolfe methods, to construct white-box adversarial attacks on the given input data. We perform a theoretical and numerical evaluation of these methods and compare them with standard approaches based on projection operations or geometrical intuition. Numerical experiments are performed on the MNIST and CIFAR-10 datasets, utilizing a multiclass logistic regression model, the convolutional neural networks (CNNs), and the Vision Transformer (ViT).
【2】Iterative Compositional Data Generation for Robot Control
标题：机器人控制的迭代合成数据生成
链接：https://arxiv.org/abs/2512.10891
摘要：收集机器人操作数据是昂贵的，这使得它不切实际的获取演示的组合大空间的任务中出现的多对象，多机器人，和多环境设置。虽然最近的生成模型可以为单个任务合成有用的数据，但它们没有利用机器人域的组成结构，并且很难推广到看不见的任务组合。我们提出了一个语义成分扩散Transformer，分解成机器人，对象，障碍物和对象特定的组件和学习他们的相互作用，通过注意。一旦在有限的任务子集上训练，我们表明我们的模型可以zero-shot生成高质量的转换，从中我们可以学习控制策略，用于看不见的任务组合。然后，我们引入了一个迭代的自我改进过程，其中合成数据通过离线强化学习进行验证，并纳入后续的训练轮。我们的方法大大提高了zero-shot性能，超过了单片和硬编码的组成基线，最终解决了几乎所有的搁置任务，并展示了有意义的组成结构的出现在学习的表示。
摘要：Collecting robotic manipulation data is expensive, making it impractical to acquire demonstrations for the combinatorially large space of tasks that arise in multi-object, multi-robot, and multi-environment settings. While recent generative models can synthesize useful data for individual tasks, they do not exploit the compositional structure of robotic domains and struggle to generalize to unseen task combinations. We propose a semantic compositional diffusion transformer that factorizes transitions into robot-, object-, obstacle-, and objective-specific components and learns their interactions through attention. Once trained on a limited subset of tasks, we show that our model can zero-shot generate high-quality transitions from which we can learn control policies for unseen task combinations. Then, we introduce an iterative self-improvement procedure in which synthetic data is validated via offline reinforcement learning and incorporated into subsequent training rounds. Our approach substantially improves zero-shot performance over monolithic and hard-coded compositional baselines, ultimately solving nearly all held-out tasks and demonstrating the emergence of meaningful compositional structure in the learned representations.
【3】Virtual camera detection: Catching video injection attacks in remote biometric systems
标题：虚拟摄像机检测：捕捉远程生物识别系统中的视频注入攻击
链接：https://arxiv.org/abs/2512.10653
摘要：面部反欺骗（FAS）是基于面部识别的远程生物特征认证系统的重要组成部分，越来越多地用于基于Web的应用程序。在新出现的威胁中，视频注入攻击--由deepfakes和虚拟摄像头软件等技术推动--对系统完整性构成了重大挑战。虽然虚拟摄像机检测（VCD）已显示出作为一种对策的潜力，但现有文献对其实际实施和评估提供了有限的见解。本研究介绍了一种基于机器学习的VCD方法，重点是它的设计和验证。该模型在与真实用户会话期间收集的元数据上进行训练。实验结果表明，该方法在识别视频注入尝试和降低恶意用户绕过FAS系统的风险方面是有效的。
摘要 ：Face anti-spoofing (FAS) is a vital component of remote biometric authentication systems based on facial recognition, increasingly used across web-based applications. Among emerging threats, video injection attacks -- facilitated by technologies such as deepfakes and virtual camera software -- pose significant challenges to system integrity. While virtual camera detection (VCD) has shown potential as a countermeasure, existing literature offers limited insight into its practical implementation and evaluation. This study introduces a machine learning-based approach to VCD, with a focus on its design and validation. The model is trained on metadata collected during sessions with authentic users. Empirical results demonstrate its effectiveness in identifying video injection attempts and reducing the risk of malicious users bypassing FAS systems.
【4】Adaptive Intrusion Detection System Leveraging Dynamic Neural Models with Adversarial Learning for 5G/6G Networks
标题：利用动态神经模型和对抗学习的5G/6 G网络自适应入侵检测系统
链接：https://arxiv.org/abs/2512.10637
备注：6 pages,2 figures, 1 Table
摘要：入侵检测系统（IDS）是保护5G/6 G网络免受内部和外部网络威胁的关键组件。虽然传统的IDS方法严重依赖于基于签名的方法，但它们很难检测到新的和不断发展的攻击。本文提出了一种先进的IDS框架，该框架利用5G/6 G网络中的对抗训练和动态神经网络，通过提供强大的实时威胁检测和响应能力来增强网络安全性。与传统的模型，这需要昂贵的再培训更新知识，该框架集成了增量学习算法，减少了频繁的再培训的需要。对抗性训练用于增强IDS对有毒数据的防御能力。通过使用更少的特征并结合统计属性，系统可以有效地检测潜在的威胁。使用NSL- KDD数据集进行的广泛评估表明，所提出的方法在抵抗数据集中毒的同时，为各种网络攻击的多类分类提供了更好的准确率（82.33%）。这项研究强调了对抗训练的动态神经网络在构建弹性IDS解决方案方面的潜力。
摘要：Intrusion Detection Systems (IDS) are critical components in safeguarding 5G/6G networks from both internal and external cyber threats. While traditional IDS approaches rely heavily on signature-based methods, they struggle to detect novel and evolving attacks. This paper presents an advanced IDS framework that leverages adversarial training and dynamic neural networks in 5G/6G networks to enhance network security by providing robust, real-time threat detection and response capabilities. Unlike conventional models, which require costly retraining to update knowledge, the proposed framework integrates incremental learning algorithms, reducing the need for frequent retraining. Adversarial training is used to fortify the IDS against poisoned data. By using fewer features and incorporating statistical properties, the system can efficiently detect potential threats. Extensive evaluations using the NSL- KDD dataset demonstrate that the proposed approach provides better accuracy of 82.33% for multiclass classification of various network attacks while resisting dataset poisoning. This research highlights the potential of adversarial-trained, dynamic neural networks for building resilient IDS solutions.
【5】UACER: An Uncertainty-Aware Critic Ensemble Framework for Robust Adversarial Reinforcement Learning
标题：UACER：一个具有不确定性的批判性集成框架，用于鲁棒对抗强化学习
链接：https://arxiv.org/abs/2512.10492
摘要：鲁棒对抗强化学习已经成为训练智能体处理真实环境中不确定干扰的有效范例，在自动驾驶和机器人控制等顺序决策领域具有重要应用。在这种范式中，代理训练通常被制定为主角和对手之间的零和马尔可夫博弈，以增强策略的鲁棒性。然而，对手的可训练性质不可避免地导致学习动态的非平稳性，导致训练不稳定性和收敛困难加剧，特别是在高维复杂环境中。在本文中，我们提出了一种新的方法，不确定性感知的批评Ensemination鲁棒对抗强化学习（UACER），它包括两个策略：1）多样化的批评集成：一组不同的K批评网络被并行利用来稳定Q值估计，而不是传统的单批评架构，以降低方差和增强鲁棒性。2)时变衰减不确定性（TDU）机制：超越简单的线性组合，我们开发了一种方差衍生的Q值聚合策略，该策略明确地结合了认知不确定性，以动态地调节探索-利用权衡，同时稳定训练过程。多个MuJoCo控制问题的综合实验验证了UACER的卓越有效性，在整体性能，稳定性和效率方面优于最先进的方法。
摘要：Robust adversarial reinforcement learning has emerged as an effective paradigm for training agents to handle uncertain disturbance in real environments, with critical applications in sequential decision-making domains such as autonomous driving and robotic control. Within this paradigm, agent training is typically formulated as a zero-sum Markov game between a protagonist and an adversary to enhance policy robustness. However, the trainable nature of the adversary inevitably induces non-stationarity in the learning dynamics, leading to exacerbated training instability and convergence difficulties, particularly in high-dimensional complex environments. In this paper, we propose a novel approach, Uncertainty-Aware Critic Ensemble for robust adversarial Reinforcement learning (UACER), which consists of two strategies: 1) Diversified critic ensemble: a diverse set of K critic networks is exploited in parallel to stabilize Q-value estimation rather than conventional single-critic architectures for both variance reduction and robustness enhancement. 2) Time-varying Decay Uncertainty (TDU) mechanism: advancing beyond simple linear combinations, we develop a variance-derived Q-value aggregation strategy that explicitly incorporates epistemic uncertainty to dynamically regulate the exploration-exploitation trade-off while simultaneously stabilizing the training process. Comprehensive experiments across several MuJoCo control problems validate the superior effectiveness of UACER, outperforming state-of-the-art methods in terms of overall performance, stability, and efficiency.
【6】The Eminence in Shadow: Exploiting Feature Boundary Ambiguity for Robust Backdoor Attacks
标题：阴影中的杰出：利用特征边界模糊性进行稳健的后门攻击
链接：https://arxiv.org/abs/2512.10402
备注：Accepted by KDD2026 Cycle 1 Research Track
摘要：深度神经网络（DNN）是关键应用的基础，但仍然容易受到后门攻击，通常依赖于启发式暴力破解方法。尽管后门研究取得了重大的经验性进展，但缺乏严格的理论分析限制了对底层机制的理解，限制了攻击的可预测性和适应性。因此，我们提供了一个针对后门攻击的理论分析，重点是稀疏决策边界如何实现不成比例的模型操作。基于这一发现，我们得到一个封闭的形式，模糊的边界区域，其中可以忽略不计的重新标记的样本诱导大量的误分类。影响函数分析进一步量化了由这些裕度样本引起的显著参数偏移，对干净准确性的影响最小，正式奠定了为什么如此低的中毒率足以进行有效攻击的基础。利用这些见解，我们提出了Eminence，一个可解释的和强大的黑盒后门框架，具有可证明的理论保证和固有的隐身特性。Eminence优化了一个通用的、视觉上微妙的触发器，该触发器战略性地利用了易受攻击的决策边界，并有效地实现了稳健的错误分类，同时具有极低的中毒率（<0.1%，而SOTA方法通常需要> 1%）。综合实验验证了我们的理论讨论，并证明了卓越的有效性，证实了边缘中毒和对抗性边界操纵之间的指数关系。Eminence保持了> 90%的攻击成功率，表现出可忽略不计的干净准确性损失，并在不同的模型，数据集和场景中表现出高度的可移植性。
摘要：Deep neural networks (DNNs) underpin critical applications yet remain vulnerable to backdoor attacks, typically reliant on heuristic brute-force methods. Despite significant empirical advancements in backdoor research, the lack of rigorous theoretical analysis limits understanding of underlying mechanisms, constraining attack predictability and adaptability. Therefore, we provide a theoretical analysis targeting backdoor attacks, focusing on how sparse decision boundaries enable disproportionate model manipulation. Based on this finding, we derive a closed-form, ambiguous boundary region, wherein negligible relabeled samples induce substantial misclassification. Influence function analysis further quantifies significant parameter shifts caused by these margin samples, with minimal impact on clean accuracy, formally grounding why such low poison rates suffice for efficacious attacks. Leveraging these insights, we propose Eminence, an explainable and robust black-box backdoor framework with provable theoretical guarantees and inherent stealth properties. Eminence optimizes a universal, visually subtle trigger that strategically exploits vulnerable decision boundaries and effectively achieves robust misclassification with exceptionally low poison rates (< 0.1%, compared to SOTA methods typically requiring > 1%). Comprehensive experiments validate our theoretical discussions and demonstrate the effectiveness of Eminence, confirming an exponential relationship between margin poisoning and adversarial boundary manipulation. Eminence maintains > 90% attack success rate, exhibits negligible clean-accuracy loss, and demonstrates high transferability across diverse models, datasets and scenarios.
【7】FLARE: A Wireless Side-Channel Fingerprinting Attack on Federated Learning
标题：DART：对联邦学习的无线侧通道指纹攻击
链接
：https://arxiv.org/abs/2512.10296
备注：This paper has been accepted for publication in IEEE INFOCOM 2026 - IEEE Conference on Computer Communications
摘要：联合学习（FL）支持跨分布式设备的协作模型训练，同时保护数据和用户隐私。然而，FL仍然容易受到隐私威胁的影响，这些威胁可能会通过直接手段损害数据。也就是说，间接地损害了FL模型架构的机密性（例如，卷积神经网络（CNN）或递归神经网络（RNN））的方法仍然未被探索。如果泄露，这些信息可以实现针对架构的下一级攻击。本文提出了一种新的侧信道指纹攻击，利用来自FL客户端的加密无线流量的流级和数据包级统计数据来推断其深度学习模型架构。我们将其命名为FLARE，这是一个基于FL体系结构识别的指纹识别框架。对各种CNN和RNN变体（包括通过IEEE 802.11 Wi-Fi训练的预训练模型和自定义模型）的评估表明，FLARE在封闭世界中的F1得分超过98%，在开放世界方案中高达91%。这些结果表明，CNN和RNN模型泄漏了可区分的流量模式，即使在具有硬件、软件和数据异构性的现实FL设置下也能实现架构指纹识别。据我们所知，这是第一个通过嗅探加密的无线流量来识别FL模型架构的工作，暴露了当前FL系统中的关键侧信道漏洞。
摘要：Federated Learning (FL) enables collaborative model training across distributed devices while safeguarding data and user privacy. However, FL remains susceptible to privacy threats that can compromise data via direct means. That said, indirectly compromising the confidentiality of the FL model architecture (e.g., a convolutional neural network (CNN) or a recurrent neural network (RNN)) on a client device by an outsider remains unexplored. If leaked, this information can enable next-level attacks tailored to the architecture. This paper proposes a novel side-channel fingerprinting attack, leveraging flow-level and packet-level statistics of encrypted wireless traffic from an FL client to infer its deep learning model architecture. We name it FLARE, a fingerprinting framework based on FL Architecture REconnaissance. Evaluation across various CNN and RNN variants-including pre-trained and custom models trained over IEEE 802.11 Wi-Fi-shows that FLARE achieves over 98% F1-score in closed-world and up to 91% in open-world scenarios. These results reveal that CNN and RNN models leak distinguishable traffic patterns, enabling architecture fingerprinting even under realistic FL settings with hardware, software, and data heterogeneity. To our knowledge, this is the first work to fingerprint FL model architectures by sniffing encrypted wireless traffic, exposing a critical side-channel vulnerability in current FL systems.
【8】Murmur2Vec: A Hashing Based Solution For Embedding Generation Of COVID-19 Spike Sequences
标题：Murmur 2 Vec：一种基于哈希的COVID-19尖峰序列嵌入生成解决方案
链接：https://arxiv.org/abs/2512.10147
摘要：由SARS-CoV-2引起的冠状病毒病（COVID-19）的早期检测和特征描述对于有效的临床应对和公共卫生规划仍然至关重要。大规模病毒序列数据的全球可用性为计算分析提供了重要机会;然而，现有方法面临着显着的局限性。基于系统发生树的方法是计算密集型的，并且不能有效地扩展到今天的数百万序列数据集。类似地，当前基于嵌入的技术通常依赖于比对序列，或者表现出次优的预测性能和高运行时间成本，从而为实际的大规模分析创造了障碍。在这项研究中，我们专注于最流行的SARS-CoV-2谱系与刺突蛋白区域，并介绍了一种可扩展的嵌入方法，利用哈希生成紧凑的，低维表示的刺突序列。这些嵌入随后被用于训练各种机器学习模型以用于监督谱系分类。我们进行了广泛的评估，将我们的方法与多个基线和最先进的生物序列嵌入方法进行比较。我们的研究结果表明，所提出的嵌入提供了显着的效率提高，实现高达86.4%的分类精度，同时减少嵌入生成时间高达99.81%。这突出了该方法作为大规模病毒序列分析的快速，有效和可扩展的解决方案的潜力。
摘要：Early detection and characterization of coronavirus disease (COVID-19), caused by SARS-CoV-2, remain critical for effective clinical response and public-health planning. The global availability of large-scale viral sequence data presents significant opportunities for computational analysis; however, existing approaches face notable limitations. Phylogenetic tree-based methods are computationally intensive and do not scale efficiently to today's multi-million-sequence datasets. Similarly, current embedding-based techniques often rely on aligned sequences or exhibit suboptimal predictive performance and high runtime costs, creating barriers to practical large-scale analysis. In this study, we focus on the most prevalent SARS-CoV-2 lineages associated with the spike protein region and introduce a scalable embedding method that leverages hashing to generate compact, low-dimensional representations of spike sequences. These embeddings are subsequently used to train a variety of machine learning models for supervised lineage classification. We conduct an extensive evaluation comparing our approach with multiple baseline and state-of-the-art biological sequence embedding methods across diverse metrics. Our results demonstrate that the proposed embeddings offer substantial improvements in efficiency, achieving up to 86.4\% classification accuracy while reducing embedding generation time by as much as 99.81\%. This highlights the method's potential as a fast, effective, and scalable solution for large-scale viral sequence analysis.
半/弱/无/有监督|不确定性|主动学习(7篇)
【1】Digital Twin Supervised Reinforcement Learning Framework for Autonomous Underwater Navigation
标题：用于自主水下导航的数字双胞胎监督强化学习框架
链接：https://arxiv.org/abs/2512.10925
摘要：由于缺乏GPS、能见度下降以及水下障碍物的存在，水下环境中的自主导航仍然是一个重大挑战。本文通过BlueROV 2（一个广泛用于科学实验的开放平台）的案例来研究这些问题。我们提出了一种基于邻近策略优化（PPO）算法的深度强化学习方法，使用一个结合了面向目标的导航信息、虚拟占用网格和沿操作区域边界的光线投射的观察空间。学习的政策进行比较，对一个参考确定性的运动规划，动态窗口方法（DWA），通常采用作为一个强大的基线避障。该评估是在真实的模拟环境中进行的，并通过在测试场地的3D数字孪生模型监督下对物理BlueROV 2进行验证来补充，有助于降低与真实世界实验相关的风险。结果表明，PPO策略在高度杂乱的环境中始终优于DWA，这主要归功于更好的本地适应和减少的冲突。最后，实验证明了学习行为从模拟到现实世界的可转移性，证实了深度RL与水下机器人自主导航的相关性。
摘要：Autonomous navigation in underwater environments remains a major challenge due to the absence of GPS, degraded visibility, and the presence of submerged obstacles. This article investigates these issues through the case of the BlueROV2, an open platform widely used for scientific experimentation. We propose a deep reinforcement learning approach based on the Proximal Policy Optimization (PPO) algorithm, using an observation space that combines target-oriented navigation information, a virtual occupancy grid, and ray-casting along the boundaries of the operational area. The learned policy is compared against a reference deterministic kinematic planner, the Dynamic Window Approach (DWA), commonly employed as a robust baseline for obstacle avoidance. The evaluation is conducted in a realistic simulation environment and complemented by validation on a physical BlueROV2 supervised by a 3D digital twin of the test site, helping to reduce risks associated with real-world experimentation. The results show that the PPO policy consistently outperforms DWA in highly cluttered environments, notably thanks to better local adaptation and reduced collisions. Finally, the experiments demonstrate the transferability of the learned behavior from simulation to the real world, confirming the relevance of deep RL for autonomous navigation in underwater robotics.
【2】Refinement Contrastive Learning of Cell-Gene Associations for Unsupervised Cell Type Identification
标题：用于无监督细胞类型识别的细胞-基因关联的细化对比学习
链接：https://arxiv.org/abs/2512.10640
摘要：在单细胞组学研究中，无监督细胞类型识别对于发现和表征异质群体至关重要。虽然已经开发了一系列的聚类方法，但大多数只关注内在的细胞结构，而忽略了细胞-基因关联的关键作用，这限制了它们区分密切相关的细胞类型的能力。为此，我们提出了一个细化对比学习框架（scRCL），明确纳入细胞基因的相互作用，以获得更多的信息表示。具体来说，我们引入了两个对比分布对齐组件，通过有效地利用细胞-细胞结构关系来揭示可靠的内在细胞结构。此外，我们开发了一个细化模块，该模块集成了基因相关结构学习，通过捕获潜在的细胞基因关联来增强细胞嵌入。该模块加强了细胞及其相关基因之间的联系，完善了表征学习，以利用生物学上有意义的关系。在几个单细胞RNA-seq和空间转录组学基准数据集上的广泛实验表明，我们的方法在细胞类型识别准确性方面始终优于最先进的基线。此外，下游生物学分析证实，回收的细胞群体表现出一致的基因表达特征，进一步验证了我们的方法的生物学相关性。该代码可在https://github.com/THPengL/scRCL上获得。
摘要：Unsupervised cell type identification is crucial for uncovering and characterizing heterogeneous populations in single cell omics studies. Although a range of clustering methods have been developed, most focus exclusively on intrinsic cellular structure and ignore the pivotal role of cell-gene associations, which limits their ability to distinguish closely related cell types. To this end, we propose a Refinement Contrastive Learning framework (scRCL) that explicitly incorporates cell-gene interactions to derive more informative representations. Specifically, we introduce two contrastive distribution alignment components that reveal reliable intrinsic cellular structures by effectively exploiting cell-cell structural relationships. Additionally, we develop a refinement module that integrates gene-correlation structure learning to enhance cell embeddings by capturing underlying cell-gene associations. This module strengthens connections between cells and their associated genes, refining the representation learning to exploiting biologically meaningful relationships. Extensive experiments on several single-cell RNA-seq and spatial transcriptomics benchmark datasets demonstrate that our method consistently outperforms state-of-the-art baselines in cell-type identification accuracy. Moreover, downstream biological analyses confirm that the recovered cell populations exhibit coherent gene-expression signatures, further validating the biological relevance of our approach. The code is available at https://github.com/THPengL/scRCL.
【3】Uncertainty-Preserving QBNNs: Multi-Level Quantization of SVI-Based Bayesian Neural Networks for Image Classification
标题：保留不确定性的QBNN：用于图像分类的基于SVI的Bayesian神经网络的多层量化
链接：https://arxiv.org/abs/2512.10602
摘要：贝叶斯神经网络（BNN）提供了原则性的不确定性量化，但与确定性网络相比，它具有大量的计算和内存开销。虽然量化技术已经成功地降低了标准深度学习模型的资源需求，但它们在概率模型中的应用在很大程度上仍未得到探索。我们介绍了一个系统的多级量化框架的随机变分推理的BNN区分三种量化策略：变分参数量化（VPQ），采样参数量化（SPQ），联合量化（JQ）。我们的对数量化的方差参数，和专门的激活函数，以保持分布结构是必不可少的校准不确定性估计。通过对Dirty-MNIST的综合实验，我们证明了BNN可以量化到4位精度，同时保持分类精度和不确定性解纠缠。在4位，联合量化实现了高达8倍的内存减少相比，浮点实现，在认知和任意不确定性估计最小的退化。这些结果使得BNN能够在资源受限的边缘设备上部署，并为未来以固有的低精度运行的模拟“贝叶斯机”提供设计指南。
摘要：Bayesian Neural Networks (BNNs) provide principled uncertainty quantification but suffer from substantial computational and memory overhead compared to deterministic networks. While quantization techniques have successfully reduced resource requirements in standard deep learning models, their application to probabilistic models remains largely unexplored. We introduce a systematic multi-level quantization framework for Stochastic Variational Inference based BNNs that distinguishes between three quantization strategies: Variational Parameter Quantization (VPQ), Sampled Parameter Quantization (SPQ), and Joint Quantization (JQ). Our logarithmic quantization for variance parameters, and specialized activation functions to preserve the distributional structure are essential for calibrated uncertainty estimation. Through comprehensive experiments on Dirty-MNIST, we demonstrate that BNNs can be quantized down to 4-bit precision while maintaining both classification accuracy and uncertainty disentanglement. At 4 bits, Joint Quantization achieves up to 8x memory reduction compared to floating-point implementations with minimal degradation in epistemic and aleatoric uncertainty estimation. These results enable deployment of BNNs on resource-constrained edge devices and provide design guidelines for future analog "Bayesian Machines" operating at inherently low precision.
【4】Hybrid Physics-ML Model for Forward Osmosis Flux with Complete Uncertainty Quantification
标题：具有完全不确定性量化的正向渗透通量混合物理-ML模型
链接：https://arxiv.org/abs/2512.10457
备注：12 pages, 6 figures
摘要：正向渗透（FO）是一种很有前途的低能膜分离技术，但由于复杂的内部传质现象，准确建模其水通量（Jw）仍然存在挑战。传统的机械模型与经验参数的可变性作斗争，而纯数据驱动的模型缺乏物理一致性和严格的不确定性量化（UQ）。这项研究介绍了一种新的鲁棒混合物理ML框架，采用高斯过程回归（GPR）进行高度准确，不确定性感知的Jw预测。核心创新在于根据详细的非线性FO物理模型预测（Jw_physical）和实验水通量（Jw_actual）之间的残差训练GPR。至关重要的是，我们通过将总预测方差（sigma2_total）分解为模型不确定性（认知的，来自GPR的后验方差）和输入不确定性（任意的，通过Delta方法分析传播的多变量相关输入）来实现完整的UQ方法。利用GPR在低数据状态下的固有优势，该模型在120个数据点上进行训练，在独立测试数据上实现了0.26%的最先进平均绝对百分比误差（MAPE）和0.999的R2，验证了先进的FO过程优化和数字孪生开发的真正强大和可靠的代理模型。
摘要：Forward Osmosis (FO) is a promising low-energy membrane separation technology, but challenges in accurately modelling its water flux (Jw) persist due to complex internal mass transfer phenomena. Traditional mechanistic models struggle with empirical parameter variability, while purely data-driven models lack physical consistency and rigorous uncertainty quantification (UQ). This study introduces a novel Robust Hybrid Physics-ML framework employing Gaussian Process Regression (GPR) for highly accurate, uncertainty-aware Jw prediction. The core innovation lies in training the GPR on the residual error between the detailed, non-linear FO physical model prediction (Jw_physical) and the experimental water flux (Jw_actual). Crucially, we implement a full UQ methodology by decomposing the total predictive variance (sigma2_total) into model uncertainty (epistemic, from GPR's posterior variance) and input uncertainty (aleatoric, analytically propagated via the Delta method for multi-variate correlated inputs). Leveraging the inherent strength of GPR in low-data regimes, the model, trained on a meagre 120 data points, achieved a state-of-the-art Mean Absolute Percentage Error (MAPE) of 0.26% and an R2 of 0.999 on the independent test data, validating a truly robust and reliable surrogate model for advanced FO process optimization and digital twin development.
【5】Solving Semi-Supervised Few-Shot Learning from an Auto-Annotation Perspective
标题：从自动注释角度解决半监督Few-Shot学习
链接：https://arxiv.org/abs/2512.10244
备注：website and code: https://tian1327.github.io/SWIFT
摘要：半监督Few-Shot学习（SSFSL）制定了现实世界的应用程序，如“自动注释”，因为它的目的是在一些标记和大量未标记的示例上学习模型来注释未标记的示例。尽管有强大的开源视觉语言模型（VLM）及其预训练数据，但SSFSL文献在很大程度上忽略了这些开源资源。相比之下，相关领域的Few-Shot学习（FSL）已经利用它们来提高性能。可以说，为了在现实世界中实现自动注释，SSFSL应该利用这些开源资源。为此，我们首先应用已建立的SSL方法来微调VLM。与直觉相反，它们的表现明显低于FSL基线。我们的深入分析揭示了根本原因：VLM产生相当“平坦”的softmax概率分布。这导致未标记数据的零利用率和弱监督信号。我们用非常简单的技术来解决这个问题：分类器初始化和温度调整。它们共同提高了伪标签的置信度，提高了未标记数据的利用率，并加强了监管信号。在此基础上，我们提出：逐步微调与温度调整（SWIFT），使现有的SSL方法能够有效地微调VLM对有限的标记数据，丰富的未标记数据，任务相关的，但噪声数据从VLM的预训练集检索。在五个SSFSL基准测试上进行的大量实验表明，SWIFT的性能比最近的FSL和SSL方法高出5个精度点。SWIFT甚至可以与监督学习相媲美，后者可以对VLM进行微调，将未标记的数据标记为地面实况！
摘要 ：Semi-supervised few-shot learning (SSFSL) formulates real-world applications like ''auto-annotation'', as it aims to learn a model over a few labeled and abundant unlabeled examples to annotate the unlabeled ones. Despite the availability of powerful open-source Vision-Language Models (VLMs) and their pretraining data, the SSFSL literature largely neglects these open-source resources. In contrast, the related area few-shot learning (FSL) has already exploited them to boost performance. Arguably, to achieve auto-annotation in the real world, SSFSL should leverage such open-source resources. To this end, we start by applying established SSL methods to finetune a VLM. Counterintuitively, they significantly underperform FSL baselines. Our in-depth analysis reveals the root cause: VLMs produce rather ''flat'' distributions of softmax probabilities. This results in zero utilization of unlabeled data and weak supervision signals. We address this issue with embarrassingly simple techniques: classifier initialization and temperature tuning. They jointly increase the confidence scores of pseudo-labels, improving the utilization rate of unlabeled data, and strengthening supervision signals. Building on this, we propose: Stage-Wise Finetuning with Temperature Tuning (SWIFT), which enables existing SSL methods to effectively finetune a VLM on limited labeled data, abundant unlabeled data, and task-relevant but noisy data retrieved from the VLM's pretraining set. Extensive experiments on five SSFSL benchmarks show that SWIFT outperforms recent FSL and SSL methods by $\sim$5 accuracy points. SWIFT even rivals supervised learning, which finetunes VLMs with the unlabeled data being labeled with ground truth!
【6】Unsupervised Acquisition of Discrete Grammatical Categories
标题：离散语法类别的无监督获取
链接：https://arxiv.org/abs/2503.18702
备注：34 pages, 3 figures, 7 tables
摘要：本文介绍了使用计算实验室环境进行语言习得实验。它实现了一个由两个代理组成的多代理系统：一个成人语言模型和一个旨在学习母语的子语言模型。至关重要的是，子代理不能访问母语言模型的内部知识，而只能访问母代理生成的语言范例。这些实验说明了如何使用该系统来获得抽象的语法知识。我们展示了如何统计分析的模式，在输入数据对应的语法类别产生离散的语法规则。这些规则随后被添加到子语言模型的语法知识中。为此，层次凝聚聚类分析应用于连续生成的母语模型的话语。有人认为，这一过程可以用来获得类似的语法类别的结构，由语言学家提出的自然语言。因此，它是建立了非平凡的语法知识已经获得。此外，使用由母语模型生成的训练数据确定的该计算实验室环境的参数配置在第二实验中用类似地导致非平凡类别的获取的测试集进行验证。
摘要：This article presents experiments performed using a computational laboratory environment for language acquisition experiments. It implements a multi-agent system consisting of two agents: an adult language model and a daughter language model that aims to learn the mother language. Crucially, the daughter agent does not have access to the internal knowledge of the mother language model but only to the language exemplars the mother agent generates. These experiments illustrate how this system can be used to acquire abstract grammatical knowledge. We demonstrate how statistical analyses of patterns in the input data corresponding to grammatical categories yield discrete grammatical rules. These rules are subsequently added to the grammatical knowledge of the daughter language model. To this end, hierarchical agglomerative cluster analysis was applied to the utterances consecutively generated by the mother language model. It is argued that this procedure can be used to acquire structures resembling grammatical categories proposed by linguists for natural languages. Thus, it is established that non-trivial grammatical knowledge has been acquired. Moreover, the parameter configuration of this computational laboratory environment determined using training data generated by the mother language model is validated in a second experiment with a test set similarly resulting in the acquisition of non-trivial categories.
【7】Supervised Learning of Random Neural Architectures Structured by Latent Random Fields on Compact Boundaryless Multiply-Connected Manifolds
标题：紧凑无边界多连通Manifold上潜随机场结构的随机神经结构的监督学习
链接：https://arxiv.org/abs/2512.10407
备注：46 pages, 12 figures
摘要：本文介绍了一种新的概率框架监督学习神经系统。它的目的是模拟复杂的，不确定的系统，其随机输出是强非高斯给定的确定性输入。该体系结构本身是一个随机对象随机生成的潜各向异性高斯随机场定义在一个紧凑的，无边界的，多连通流形。我们的目标是建立一个新的概念和数学框架，其中神经架构是几何感知，场驱动的生成过程的实现。神经拓扑结构和突触权重都来自一个潜在的随机场。降阶参数化管理的空间强度的非齐次泊松过程的流形上，从神经元的位置进行采样。输入和输出神经元通过隐场的极值评估来识别，而连接性通过测地线邻近和局部场亲和来建立。突触权重有条件地从现场实现采样，诱导随机输出响应，即使是确定性的输入。为了确保可扩展性，该架构是稀疏通过基于网格的扩散掩蔽，产生几何感知稀疏连接，没有特设的结构假设。监督学习被公式化为对潜在场的生成超参数的推断，使用通过蒙特卡洛采样从单观测每输入数据集估计的负对数似然损失。本文开始了数学分析的模型，建立基本属性，如适定性，可测性，并诱导随机映射，支持其内部的一致性，奠定了基础，为更广泛的几何驱动的随机学习理论的表达变异的初步分析。
摘要：This paper introduces a new probabilistic framework for supervised learning in neural systems. It is designed to model complex, uncertain systems whose random outputs are strongly non-Gaussian given deterministic inputs. The architecture itself is a random object stochastically generated by a latent anisotropic Gaussian random field defined on a compact, boundaryless, multiply-connected manifold. The goal is to establish a novel conceptual and mathematical framework in which neural architectures are realizations of a geometry-aware, field-driven generative process. Both the neural topology and synaptic weights emerge jointly from a latent random field. A reduced-order parameterization governs the spatial intensity of an inhomogeneous Poisson process on the manifold, from which neuron locations are sampled. Input and output neurons are identified via extremal evaluations of the latent field, while connectivity is established through geodesic proximity and local field affinity. Synaptic weights are conditionally sampled from the field realization, inducing stochastic output responses even for deterministic inputs. To ensure scalability, the architecture is sparsified via percentile-based diffusion masking, yielding geometry-aware sparse connectivity without ad hoc structural assumptions. Supervised learning is formulated as inference on the generative hyperparameters of the latent field, using a negative log-likelihood loss estimated through Monte Carlo sampling from single-observation-per-input datasets. The paper initiates a mathematical analysis of the model, establishing foundational properties such as well-posedness, measurability, and a preliminary analysis of the expressive variability of the induced stochastic mappings, which support its internal coherence and lay the groundwork for a broader theory of geometry-driven stochastic learning.
迁移|Zero/Few/One-Shot|自适应(6篇)
【1】Guided Transfer Learning for Discrete Diffusion Models
标题：离散扩散模型的引导迁移学习
链接：https://arxiv.org/abs/2512.10877
备注：7 pages (main text) + appendix
摘要：离散扩散模型在语言和其他离散域中具有很强的性能，为自回归模型提供了一个强大的替代方案。然而，它们的强大性能依赖于大型训练数据集，这是昂贵的或有风险的，特别是在适应新领域时。迁移学习是适应预训练离散扩散模型的自然方法，但目前的方法需要对大型扩散模型进行微调，这在计算上是昂贵的，而且往往不切实际。基于连续扩散的基于比率的迁移学习，我们为离散扩散模型（GTL）提供了引导迁移学习。这使得能够在不修改预训练的去噪器的情况下从目标分布进行采样。相同的指导公式适用于离散时间扩散和基于连续时间分数的离散扩散，从而得到统一的处理。引导离散扩散通常需要引导网络的许多前向传递，这对于大词汇表和长序列变得不切实际。为了解决这个问题，我们进一步提出了一个有效的引导采样器，集中评估计划者选择的位置和顶部候选令牌，从而降低采样时间和计算。这使得引导式语言建模在大规模词汇表和长序列中是实用的。我们评估GTL序列数据，包括合成马尔可夫链和语言建模，并提供其行为的实证分析。
摘要 ：Discrete diffusion models achieve strong performance across language and other discrete domains, providing a powerful alternative to autoregressive models. However, their strong performance relies on large training datasets, which are costly or risky to obtain, especially when adapting to new domains. Transfer learning is the natural way to adapt pretrained discrete diffusion models, but current methods require fine-tuning large diffusion models, which is computationally expensive and often impractical. Building on ratio-based transfer learning for continuous diffusion, we provide Guided Transfer Learning for discrete diffusion models (GTL). This enables sampling from a target distribution without modifying the pretrained denoiser. The same guidance formulation applies to both discrete-time diffusion and continuous-time score-based discrete diffusion, yielding a unified treatment. Guided discrete diffusion often requires many forward passes of the guidance network, which becomes impractical for large vocabularies and long sequences. To address this, we further present an efficient guided sampler that concentrates evaluations on planner-selected positions and top candidate tokens, thus lowering sampling time and computation. This makes guided language modeling practical at scale for large vocabularies and long sequences. We evaluate GTL on sequential data, including synthetic Markov chains and language modeling, and provide empirical analyses of its behavior.
【2】Grow Up and Merge: Scaling Strategies for Efficient Language Adaptation
标题：成长与合并：高效语言适应的扩展策略
链接：https://arxiv.org/abs/2512.10772
摘要：实现包括中等和低资源语言的高性能语言模型仍然是一个挑战。与特定语言的适应相比，大规模多语言模型仍然表现不佳，特别是在较小的模型规模下。在这项工作中，我们研究了缩放作为一种有效的策略，用于使预训练模型适应新的目标语言。通过使用近似FLOP匹配的模型进行全面的缩放消融，我们测试了英语基础模型的升级是否比标准的持续预训练更有效，更节省资源。我们发现，一旦暴露于足够的目标语言数据，较大的升级模型可以匹配或超过在更多数据上持续预训练的较小模型的性能，这证明了扩展对数据效率的好处。缩放还有助于保留基本模型的英语功能，从而减少灾难性的遗忘。最后，我们探讨这种规模化的，特定语言的模型是否可以合并，以构建模块化和灵活的多语言系统。我们发现，虽然合并仍然不如联合多语言训练有效，但升级的合并比小型合并表现更好。我们观察到大的性能差异合并方法，建议通过合并方法专门用于语言级集成的改进潜力。
摘要：Achieving high-performing language models which include medium- and lower-resource languages remains a challenge. Massively multilingual models still underperform compared to language-specific adaptations, especially at smaller model scales. In this work, we investigate scaling as an efficient strategy for adapting pretrained models to new target languages. Through comprehensive scaling ablations with approximately FLOP-matched models, we test whether upscaling an English base model enables more effective and resource-efficient adaptation than standard continued pretraining. We find that, once exposed to sufficient target-language data, larger upscaled models can match or surpass the performance of smaller models continually pretrained on much more data, demonstrating the benefits of scaling for data efficiency. Scaling also helps preserve the base model's capabilities in English, thus reducing catastrophic forgetting. Finally, we explore whether such scaled, language-specific models can be merged to construct modular and flexible multilingual systems. We find that while merging remains less effective than joint multilingual training, upscaled merges perform better than smaller ones. We observe large performance differences across merging methods, suggesting potential for improvement through merging approaches specialized for language-level integration.
【3】Adaptive Replay Buffer for Offline-to-Online Reinforcement Learning
标题：用于离线到在线强化学习的自适应重播缓冲区
链接：https://arxiv.org/abs/2512.10510
备注：15 pages, 3 figures, 7 tables
摘要：离线到在线强化学习（O2O RL）在平衡使用固定的离线数据集与新收集的在线体验方面面临着一个关键的困境。标准方法通常依赖于固定的数据混合比率，难以在早期学习稳定性和渐近性能之间进行权衡。为了克服这一点，我们引入了自适应重放缓冲区（ARB），一种新的方法，动态优先数据采样的基础上，我们称之为“上policyness”的轻量级指标。与依赖复杂学习过程或固定比率的现有方法不同，ARB的设计无需学习，易于实现，可无缝集成到现有的O2O RL算法中。它评估收集的轨迹与当前策略行为的紧密程度，并为该轨迹内的每个转换分配比例采样权重。这一策略有效地利用了离线数据来实现初始稳定性，同时逐步将学习重点放在最相关、最有价值的在线体验上。我们在D4RL基准测试上进行的大量实验表明，ARB始终可以缓解早期性能下降，并显着提高各种O2O RL算法的最终性能，突出了自适应，行为感知重放缓冲区设计的重要性。
摘要：Offline-to-Online Reinforcement Learning (O2O RL) faces a critical dilemma in balancing the use of a fixed offline dataset with newly collected online experiences. Standard methods, often relying on a fixed data-mixing ratio, struggle to manage the trade-off between early learning stability and asymptotic performance. To overcome this, we introduce the Adaptive Replay Buffer (ARB), a novel approach that dynamically prioritizes data sampling based on a lightweight metric we call 'on-policyness'. Unlike prior methods that rely on complex learning procedures or fixed ratios, ARB is designed to be learning-free and simple to implement, seamlessly integrating into existing O2O RL algorithms. It assesses how closely collected trajectories align with the current policy's behavior and assigns a proportional sampling weight to each transition within that trajectory. This strategy effectively leverages offline data for initial stability while progressively focusing learning on the most relevant, high-rewarding online experiences. Our extensive experiments on D4RL benchmarks demonstrate that ARB consistently mitigates early performance degradation and significantly improves the final performance of various O2O RL algorithms, highlighting the importance of an adaptive, behavior-aware replay buffer design.
【4】R^2-HGP: A Double-Regularized Gaussian Process for Heterogeneous Transfer Learning
标题：R#2-HGP：一种用于异类迁移学习的双正规高斯过程
链接：https://arxiv.org/abs/2512.10258
备注：17 pages, 9 figures. Under review for IEEE TPAMI
摘要：多输出高斯过程（MGP）模型因其灵活性和不确定性量化能力而受到广泛关注，并因其捕获任务间相关性的能力而被广泛应用于多源迁移学习场景。然而，他们在迁移学习中仍然面临着一些挑战。首先，源域和目标域的输入空间往往是异质的，这使得直接的知识转移变得困难。其次，潜在的先验知识和物理信息通常被忽略在异构传输，阻碍了特定领域的见解的利用，并导致不稳定的映射。第三，目标和来源之间不适当的信息共享容易导致负迁移。传统模式无法以统一的方式解决这些问题。为了克服这些局限性，本文提出了一种双正则化异质高斯过程框架（R^2-HGP）。具体而言，首先提出了一种可训练的先验概率映射模型来对齐异构输入域。将得到的对齐输入视为潜在变量，在此基础上构建多源传输GP模型，并将整个结构集成到一个新的基于条件变分自编码器（CVAE）的框架中。物理洞察力被进一步并入作为正则化项，以确保对齐结果坚持已知的物理知识。接下来，在多源传输GP模型中，对传输系数施加稀疏惩罚，使得模型能够自适应地选择信息量最大的源输出并抑制负传输。广泛的模拟和真实世界的工程案例研究验证了我们的R^2-HGP的有效性，在不同的评估指标上表现出了与最先进的基准测试相比的一致优势。
摘要 ：Multi-output Gaussian process (MGP) models have attracted significant attention for their flexibility and uncertainty-quantification capabilities, and have been widely adopted in multi-source transfer learning scenarios due to their ability to capture inter-task correlations. However, they still face several challenges in transfer learning. First, the input spaces of the source and target domains are often heterogeneous, which makes direct knowledge transfer difficult. Second, potential prior knowledge and physical information are typically ignored during heterogeneous transfer, hampering the utilization of domain-specific insights and leading to unstable mappings. Third, inappropriate information sharing among target and sources can easily lead to negative transfer. Traditional models fail to address these issues in a unified way. To overcome these limitations, this paper proposes a Double-Regularized Heterogeneous Gaussian Process framework (R^2-HGP). Specifically, a trainable prior probability mapping model is first proposed to align the heterogeneous input domains. The resulting aligned inputs are treated as latent variables, upon which a multi-source transfer GP model is constructed and the entire structure is integrated into a novel conditional variational autoencoder (CVAE) based framework. Physical insights is further incorporated as a regularization term to ensure that the alignment results adhere to known physical knowledge. Next, within the multi-source transfer GP model, a sparsity penalty is imposed on the transfer coefficients, enabling the model to adaptively select the most informative source outputs and suppress negative transfer. Extensive simulations and real-world engineering case studies validate the effectiveness of our R^2-HGP, demonstrating consistent superiority over state-of-the-art benchmarks across diverse evaluation metrics.
【5】Adaptive Information Routing for Multimodal Time Series Forecasting
标题：多峰时间序列预测的自适应信息路由
链接：https://arxiv.org/abs/2512.10229
摘要：时间序列预测是人工智能的一项重要任务，具有许多实际应用。传统的方法主要依靠历史时间序列数据来预测未来值。然而，在实际情况下，由于可用信息有限，这通常不足以进行准确的预测。为了应对这一挑战，已经探索了多模式时间序列预测方法，该方法结合了额外的数据模式，主要是文本数据以及时间序列数据。在这项工作中，我们介绍了自适应信息路由（AIR）框架，多模态时间序列预测的一种新方法。与将文本数据与时间序列数据同等对待作为预测的可互换辅助特征的现有方法不同，AIR利用文本信息通过控制多变量时间序列信息应该如何组合以及组合到什么程度来动态地引导时间序列模型。我们还提出了一个文本细化管道，采用大型语言模型将原始文本数据转换为适合多模态预测的形式，我们引入了一个基准，促进基于此管道的多模态预测实验。实验结果与真实世界的市场数据，如原油价格和汇率表明，AIR有效地调节使用文本输入的时间序列模型的行为，显着提高预测精度在各种时间序列预测任务。
摘要：Time series forecasting is a critical task for artificial intelligence with numerous real-world applications. Traditional approaches primarily rely on historical time series data to predict the future values. However, in practical scenarios, this is often insufficient for accurate predictions due to the limited information available. To address this challenge, multimodal time series forecasting methods which incorporate additional data modalities, mainly text data, alongside time series data have been explored. In this work, we introduce the Adaptive Information Routing (AIR) framework, a novel approach for multimodal time series forecasting. Unlike existing methods that treat text data on par with time series data as interchangeable auxiliary features for forecasting, AIR leverages text information to dynamically guide the time series model by controlling how and to what extent multivariate time series information should be combined. We also present a text-refinement pipeline that employs a large language model to convert raw text data into a form suitable for multimodal forecasting, and we introduce a benchmark that facilitates multimodal forecasting experiments based on this pipeline. Experiment results with the real world market data such as crude oil price and exchange rates demonstrate that AIR effectively modulates the behavior of the time series model using textual inputs, significantly enhancing forecasting accuracy in various time series forecasting tasks.
【6】Inference for Batched Adaptive Experiments
标题：批量自适应实验的推断
链接：https://arxiv.org/abs/2512.10156
摘要：适应性实验的优势使其迅速被经济学、其他领域以及从业者所采用。然而，适应性实验对因果推理提出了挑战。本文提出了一种BOLS（batched ordinary least squares）检验统计量，用于自适应实验中处理效应的推断。该统计量提供了异方差性下各阶段给药-对照差异的精密度均衡汇总。组合检验统计量是每周期异方差z统计量的归一化平均值，可用于构造渐近有效的置信区间。我们提供了模拟结果，比较了典型情况下的拒收率，每批处理周期很少（或很多）。
摘要：The advantages of adaptive experiments have led to their rapid adoption in economics, other fields, as well as among practitioners. However, adaptive experiments pose challenges for causal inference. This note suggests a BOLS (batched ordinary least squares) test statistic for inference of treatment effects in adaptive experiments. The statistic provides a precision-equalizing aggregation of per-period treatment-control differences under heteroskedasticity. The combined test statistic is a normalized average of heteroskedastic per-period z-statistics and can be used to construct asymptotically valid confidence intervals. We provide simulation results comparing rejection rates in the typical case with few treatment periods and few (or many) observations per batch.
强化学习(1篇)
【1】Curriculum-Based Reinforcement Learning for Autonomous UAV Navigation in Unknown Curved Tubular Conduit
标题：未知弯曲管状管道中无人机自主导航基于课程的强化学习
链接：https://arxiv.org/abs/2512.10934
摘要：由于管道的约束几何形状、壁的接近度以及此类场景固有的感知限制，在受限管状环境中的自主无人机导航仍然是一个重大挑战。我们提出了一种强化学习方法，使无人机能够在没有任何几何形状先验知识的情况下导航未知的三维管道，仅依赖于激光雷达的局部观测和管道中心的条件视觉检测。相比之下，作为确定性基线的Pure Pursuit算法受益于对中心线的显式访问，从而创建了一种信息不对称，旨在评估RL补偿几何模型缺失的能力。智能体通过渐进的课程学习策略进行训练，逐渐将其暴露于越来越弯曲的几何形状，其中管中心经常从视野中消失。转向协商机制，直接可见性，方向记忆和激光雷达对称线索的组合的基础上，证明是必不可少的，以确保稳定的导航在这样的部分可观测性条件下。实验表明，PPO策略获得了强大的和可推广的行为，始终优于确定性控制器，尽管其有限的几何信息的访问。在高保真3D环境中的验证进一步证实了学习行为到连续物理动力学的可转移性。 因此，所提出的方法提供了一个完整的框架，在未知的管状环境中自主导航，并打开工业，地下，或医疗应用的前景，通过狭窄和弱感知管道的进展代表了一个核心挑战。
摘要：Autonomous drone navigation in confined tubular environments remains a major challenge due to the constraining geometry of the conduits, the proximity of the walls, and the perceptual limitations inherent to such scenarios. We propose a reinforcement learning approach enabling a drone to navigate unknown three-dimensional tubes without any prior knowledge of their geometry, relying solely on local observations from LiDAR and a conditional visual detection of the tube center. In contrast, the Pure Pursuit algorithm, used as a deterministic baseline, benefits from explicit access to the centerline, creating an information asymmetry designed to assess the ability of RL to compensate for the absence of a geometric model. The agent is trained through a progressive Curriculum Learning strategy that gradually exposes it to increasingly curved geometries, where the tube center frequently disappears from the visual field. A turning-negotiation mechanism, based on the combination of direct visibility, directional memory, and LiDAR symmetry cues, proves essential for ensuring stable navigation under such partial observability conditions. Experiments show that the PPO policy acquires robust and generalizable behavior, consistently outperforming the deterministic controller despite its limited access to geometric information. Validation in a high-fidelity 3D environment further confirms the transferability of the learned behavior to a continuous physical dynamics. The proposed approach thus provides a complete framework for autonomous navigation in unknown tubular environments and opens perspectives for industrial, underground, or medical applications where progressing through narrow and weakly perceptive conduits represents a central challenge.
符号|符号学习(1篇)
【1】Bayesian Symbolic Regression via Posterior Sampling
标题：通过后验抽样的Bayesian符号回归
链接：https://arxiv.org/abs/2512.10849
摘要：符号回归是直接从数据中发现控制方程的有力工具，但其对噪声的敏感性阻碍了其更广泛的应用。本文介绍了一种用于贝叶斯符号回归的序贯蒙特卡罗（SMC）框架，该框架近似符号表达式的后验分布，增强了鲁棒性，并在存在噪声的情况下实现了符号回归的不确定性量化。不同于传统的遗传编程方法，基于SMC的算法结合概率选择，自适应回火，并使用归一化的边缘似然有效地探索符号表达式的搜索空间，产生简约的表达式与改进的泛化。与标准遗传编程基线相比，所提出的方法更好地处理具有挑战性的噪声基准数据集。减少过度拟合的趋势和增强发现准确和可解释的方程的能力，为科学发现和工程设计应用中更强大的符号回归铺平了道路。
摘要：Symbolic regression is a powerful tool for discovering governing equations directly from data, but its sensitivity to noise hinders its broader application. This paper introduces a Sequential Monte Carlo (SMC) framework for Bayesian symbolic regression that approximates the posterior distribution over symbolic expressions, enhancing robustness and enabling uncertainty quantification for symbolic regression in the presence of noise. Differing from traditional genetic programming approaches, the SMC-based algorithm combines probabilistic selection, adaptive tempering, and the use of normalized marginal likelihood to efficiently explore the search space of symbolic expressions, yielding parsimonious expressions with improved generalization. When compared to standard genetic programming baselines, the proposed method better deals with challenging, noisy benchmark datasets. The reduced tendency to overfit and enhanced ability to discover accurate and interpretable equations paves the way for more robust symbolic regression in scientific discovery and engineering design applications.
分层学习(1篇)
【1】Clustered Federated Learning with Hierarchical Knowledge Distillation
标题：采用分层知识提炼的分层联邦学习
链接：https://arxiv.org/abs/2512.10443
摘要：分布式联合学习（CFL）已经成为解决数据异构性和确保大型分布式物联网环境中隐私的强大方法。通过对客户端进行聚类并训练特定于聚类的模型，CFL可以为异构客户端组定制个性化模型。然而，传统的CFL方法遭受碎片化的学习，为每个集群训练独立的全局模型，并未能利用集体集群的见解。本文提倡转向分层CFL，允许两级聚合在边缘训练特定于集群的模型，并在云中建立统一的全局模型。这种转变提高了培训效率，但可能会带来沟通方面的挑战。为此，我们提出了CFLHKD，一种新的个性化方案，将层次聚类知识集成到CFL。建立在多教师知识蒸馏，CFLHKD使集群间的知识共享，同时保留集群特定的个性化。CFLHKD采用双层聚合，以弥合本地和全球学习之间的差距。对标准基准数据集的广泛评估表明，CFLHKD在集群特定和全局模型精度方面优于代表性基线，并实现了3.32- 7.57%的性能改进。
摘要：Clustered Federated Learning (CFL) has emerged as a powerful approach for addressing data heterogeneity and ensuring privacy in large distributed IoT environments. By clustering clients and training cluster-specific models, CFL enables personalized models tailored to groups of heterogeneous clients. However, conventional CFL approaches suffer from fragmented learning for training independent global models for each cluster and fail to take advantage of collective cluster insights. This paper advocates a shift to hierarchical CFL, allowing bi-level aggregation to train cluster-specific models at the edge and a unified global model at the cloud. This shift improves training efficiency yet might introduce communication challenges. To this end, we propose CFLHKD, a novel personalization scheme for integrating hierarchical cluster knowledge into CFL. Built upon multi-teacher knowledge distillation, CFLHKD enables inter-cluster knowledge sharing while preserving cluster-specific personalization. CFLHKD adopts a bi-level aggregation to bridge the gap between local and global learning. Extensive evaluations of standard benchmark datasets demonstrate that CFLHKD outperforms representative baselines in cluster-specific and global model accuracy and achieves a performance improvement of 3.32-7.57\%.
医学相关(2篇)
【1】AutoMedic: An Automated Evaluation Framework for Clinical Conversational Agents with Medical Dataset Grounding
标题：AutoMedic：具有医疗数据集基础的临床对话代理自动评估框架
链接：https://arxiv.org/abs/2512.10195
摘要：评估大型语言模型（LLM）最近已经成为LLM在医学领域安全和可信应用的关键问题。虽然已经提出了各种静态的医疗问答（QA）基准，许多方面仍有待探索，如在动态的，互动的临床多轮对话的情况下，多方面的评估策略，超越简单的准确性识别LLM在生成响应的有效性。然而，正式评估动态、交互式临床情况受到可能的患者状态和交互轨迹的巨大组合空间的阻碍，使得难以标准化和定量测量此类场景。在这里，我们介绍了AutoMedic，一个多智能体模拟框架，可以自动评估LLM作为临床会话代理。AutoMedic将现成的静态QA数据集转换为虚拟患者配置文件，使LLM代理之间的现实和临床接地多轮临床对话成为可能。然后，基于我们的CARE度量标准评估各种临床会话代理的性能，该度量标准提供了临床会话准确性、效率/策略、同理心和鲁棒性的多方面评估标准。我们的研究结果，由人类专家验证，证明了AutoMedic作为临床会话代理的自动化评估框架的有效性，为会话医学应用中LLM的有效开发提供了实用指南。
摘要：Evaluating large language models (LLMs) has recently emerged as a critical issue for safe and trustworthy application of LLMs in the medical domain. Although a variety of static medical question-answering (QA) benchmarks have been proposed, many aspects remain underexplored, such as the effectiveness of LLMs in generating responses in dynamic, interactive clinical multi-turn conversation situations and the identification of multi-faceted evaluation strategies beyond simple accuracy. However, formally evaluating a dynamic, interactive clinical situation is hindered by its vast combinatorial space of possible patient states and interaction trajectories, making it difficult to standardize and quantitatively measure such scenarios. Here, we introduce AutoMedic, a multi-agent simulation framework that enables automated evaluation of LLMs as clinical conversational agents. AutoMedic transforms off-the-shelf static QA datasets into virtual patient profiles, enabling realistic and clinically grounded multi-turn clinical dialogues between LLM agents. The performance of various clinical conversational agents is then assessed based on our CARE metric, which provides a multi-faceted evaluation standard of clinical conversational accuracy, efficiency/strategy, empathy, and robustness. Our findings, validated by human experts, demonstrate the validity of AutoMedic as an automated evaluation framework for clinical conversational agents, offering practical guidelines for the effective development of LLMs in conversational medical applications.
【2】MedXAI: A Retrieval-Augmented and Self-Verifying Framework for Knowledge-Guided Medical Image Analysis
标题：MedXAI：一个用于知识引导医学图像分析的检索增强和自定义框架
链接：https://arxiv.org/abs/2512.10098
备注：https://cmsworkshops.com/Asilomar2025/Papers/Uploads/FinalPapers/Original/1527/20251130102314_899554_1527.pdf
摘要 ：准确且可解释的基于图像的诊断仍然是医疗人工智能的根本挑战，特别是在领域转变和罕见情况下。深度学习模型经常与现实世界的分布变化作斗争，对罕见的病理表现出偏见，并且缺乏在安全关键的临床环境中部署所需的透明度。我们引入MedXAI（医学成像分类的可解释框架），这是一个统一的基于专家知识的框架，它将深度视觉模型与临床医生获得的专家知识相结合，以提高泛化能力，减少罕见类别偏倚，并通过定位相关诊断特征而不是依赖于技术事后方法（例如，显著性图，LIME）。我们在两项具有挑战性的任务上评估了MedXAI在异质模式中的作用：（i）静息状态fMRI的癫痫发作区定位，以及（ii）糖尿病视网膜病变分级。在10个多中心数据集上的实验显示出一致的收益，包括跨域泛化提高3%，稀有类F1得分提高10%，大大优于强大的深度学习基线。消融证实，符号组件作为有效的临床先验和正则化，提高鲁棒性下的分布偏移。MedXAI提供临床一致的解释，同时实现卓越的域内和跨域性能，特别是对于多模式医疗AI中的罕见疾病。
摘要：Accurate and interpretable image-based diagnosis remains a fundamental challenge in medical AI, particularly un- der domain shifts and rare-class conditions. Deep learning mod- els often struggle with real-world distribution changes, exhibit bias against infrequent pathologies, and lack the transparency required for deployment in safety-critical clinical environments. We introduce MedXAI (An Explainable Framework for Med- ical Imaging Classification), a unified expert knowledge based framework that integrates deep vision models with clinician- derived expert knowledge to improve generalization, reduce rare- class bias, and provide human-understandable explanations by localizing the relevant diagnostic features rather than relying on technical post-hoc methods (e.g., Saliency Maps, LIME). We evaluate MedXAI across heterogeneous modalities on two challenging tasks: (i) Seizure Onset Zone localization from resting-state fMRI, and (ii) Diabetic Retinopathy grading. Ex periments on ten multicenter datasets show consistent gains, including a 3% improvement in cross-domain generalization and a 10% improvmnet in F1 score of rare class, substantially outperforming strong deep learning baselines. Ablations confirm that the symbolic components act as effective clinical priors and regularizers, improving robustness under distribution shift. MedXAI delivers clinically aligned explanations while achieving superior in-domain and cross-domain performance, particularly for rare diseases in multimodal medical AI.
蒸馏|知识提取(1篇)
【1】Hyperspectral Image Data Reduction for Endmember Extraction
标题：用于端元提取的高光谱图像数据缩减
链接：https://arxiv.org/abs/2512.10506
摘要：从高光谱图像中提取端元的目的是识别存在于场景中的物质的光谱特征。最近的研究表明，自字典方法可以实现高的提取精度，但是，其高计算成本限制了其适用于大规模的高光谱图像。虽然已经提出了几种方法来缓解这个问题，但它仍然是一个重大挑战。出于这种情况下，本文追求的数据减少的方法。假设高光谱图像遵循纯像素假设的线性混合模型，我们开发了一种数据简化技术，去除不包含端元的像素。我们分析了这个减少步骤的理论特性，并表明它保留了靠近端元的像素。在此基础上，我们提出了一个数据减少的自字典方法，它集成了数据减少与基于线性规划公式的自字典方法。数值实验表明，该方法可以大大减少计算时间的原始自字典方法，而不牺牲端元提取精度。
摘要：Endmember extraction from hyperspectral images aims to identify the spectral signatures of materials present in a scene. Recent studies have shown that self-dictionary methods can achieve high extraction accuracy; however, their high computational cost limits their applicability to large-scale hyperspectral images. Although several approaches have been proposed to mitigate this issue, it remains a major challenge. Motivated by this situation, this paper pursues a data reduction approach. Assuming that the hyperspectral image follows the linear mixing model with the pure-pixel assumption, we develop a data reduction technique that removes pixels that do not contain endmembers. We analyze the theoretical properties of this reduction step and show that it preserves pixels that lie close to the endmembers. Building on this result, we propose a data-reduced self-dictionary method that integrates the data reduction with a self-dictionary method based on a linear programming formulation. Numerical experiments demonstrate that the proposed method can substantially reduce the computational time of the original self-dictionary method without sacrificing endmember extraction accuracy.
推荐(1篇)
【1】STARS: Semantic Tokens with Augmented Representations for Recommendation at Scale
标题：STARS：具有增强表示的语义代币，可供大规模推荐
链接：https://arxiv.org/abs/2512.10149
摘要：现实世界的电子商务推荐系统必须在严格的几十毫秒延迟限制下提供相关项目，尽管存在诸如冷启动产品，快速变化的用户意图以及包括季节性，假期和促销在内的动态上下文等挑战。我们引入了STARS，这是一个基于transformer的顺序推荐框架，专为大规模，低延迟的电子商务环境而构建。STARS结合了几项创新：将长期偏好与短期会话意图分离的双记忆用户嵌入;融合预训练文本嵌入、可学习增量和LLM派生属性标签的语义项标记，增强基于内容的匹配、长尾覆盖和冷启动性能;使用学习的日历和事件偏移量进行上下文感知评分;以及一个延迟敏感的两阶段检索管道，它执行离线嵌入生成和在线最大内积搜索，并进行过滤，从而实现数十毫秒的响应时间。在对生产规模数据的离线评估中，相对于我们现有的LambdaMART系统，STARS将Hit@5提高了75%以上。对600万次访问的大规模A/B测试显示了统计学上的显著提升，包括总订单+0.8%，主页上的添加到购物车+2.0%，以及每个用户的访问量+0.5%。这些结果表明，结合语义丰富，多意图建模和面向部署的设计可以在现实环境中产生最先进的推荐质量，而不会牺牲服务效率。
摘要：Real-world ecommerce recommender systems must deliver relevant items under strict tens-of-milliseconds latency constraints despite challenges such as cold-start products, rapidly shifting user intent, and dynamic context including seasonality, holidays, and promotions. We introduce STARS, a transformer-based sequential recommendation framework built for large-scale, low-latency ecommerce settings. STARS combines several innovations: dual-memory user embeddings that separate long-term preferences from short-term session intent; semantic item tokens that fuse pretrained text embeddings, learnable deltas, and LLM-derived attribute tags, strengthening content-based matching, long-tail coverage, and cold-start performance; context-aware scoring with learned calendar and event offsets; and a latency-conscious two-stage retrieval pipeline that performs offline embedding generation and online maximum inner-product search with filtering, enabling tens-of-milliseconds response times. In offline evaluations on production-scale data, STARS improves Hit@5 by more than 75 percent relative to our existing LambdaMART system. A large-scale A/B test on 6 million visits shows statistically significant lifts, including Total Orders +0.8%, Add-to-Cart on Home +2.0%, and Visits per User +0.5%. These results demonstrate that combining semantic enrichment, multi-intent modeling, and deployment-oriented design can yield state-of-the-art recommendation quality in real-world environments without sacrificing serving efficiency.
聚类(2篇)
【1】CIEGAD: Cluster-Conditioned Interpolative and Extrapolative Framework for Geometry-Aware and Domain-Aligned Data Augmentation
标题：CIEGAD：用于几何感知和领域对齐数据增强的复杂条件内插和外推框架
链接：https://arxiv.org/abs/2512.10178
摘要：在实际的深度学习部署中，数据的稀缺性和标签分布的不平衡往往会导致真实世界数据分布中的语义未覆盖区域，阻碍模型训练，并导致类边界附近的错误分类以及外围区域的不稳定行为。尽管最近的大型语言模型（LLM）显示出数据增强的前景，但同时实现生成方向控制、域对齐和质量控制的集成框架尚未完全建立。为了应对这些挑战，我们提出了一个几何感知和域对齐数据增强（CIEGAD）的条件插值和外推框架，该框架系统地补充了分布和分布外的语义未覆盖区域。CIEGAD通过聚类条件化构造领域轮廓，采用综合类频率和几何指标的层次频率-几何分配方法进行生成分配，通过内插和外推合成的共存精细控制生成方向。它还通过几何约束过滤与LLM作为判断机制相结合来执行质量控制。多个分类任务上的实验表明，CIEGAD有效地扩展了真实世界数据分布的外围，同时保持了生成数据与真实世界数据之间的高度一致性以及语义多样性。特别是，对于长尾和多类分类任务，CIEGAD不断提高F1和召回率，验证了分布一致性、多样性和质量的三重和谐。这些结果表明，CIEGAD作为一个面向实际的数据增强框架，补充代表性不足的地区，同时保持与现实世界的数据一致。
摘要 ：In practical deep learning deployment, the scarcity of data and the imbalance of label distributions often lead to semantically uncovered regions within the real-world data distribution, hindering model training and causing misclassification near class boundaries as well as unstable behaviors in peripheral areas. Although recent large language models (LLMs) show promise for data augmentation, an integrated framework that simultaneously achieves directional control of generation, domain alignment, and quality control has not yet been fully established. To address these challenges, we propose a Cluster-conditioned Interpolative and Extrapolative framework for Geometry-Aware and Domain-aligned data augmentation (CIEGAD), which systematically complements both in-distribution and out-of-distribution semantically uncovered regions. CIEGAD constructs domain profiles through cluster conditioning, allocates generation with a hierarchical frequency-geometric allocation integrating class frequency and geometric indicators, and finely controls generation directions via the coexistence of interpolative and extrapolative synthesis. It further performs quality control through geometry-constrained filtering combined with an LLM-as-a-Judge mechanism. Experiments on multiple classification tasks demonstrate that CIEGAD effectively extends the periphery of real-world data distributions while maintaining high alignment between generated and real-world data as well as semantic diversity. In particular, for long-tailed and multi-class classification tasks, CIEGAD consistently improves F1 and recall, validating the triple harmony of distributional consistency, diversity, and quality. These results indicate that CIEGAD serves as a practically oriented data augmentation framework that complements underrepresented regions while preserving alignment with real-world data.
【2】Cluster-Dags as Powerful Background Knowledge For Causal Discovery
标题：Chicker-Dags是发现因果关系的强大背景知识
链接：https://arxiv.org/abs/2512.10032
备注：23 pages, 5 figures
摘要：寻找因果关系在科学中至关重要。因果发现的目的是从数据中恢复一个图表，简洁地描述这些因果关系。然而，目前的方法面临着一些挑战，特别是在处理高维数据和复杂的依赖关系。阐明系统的先验知识有助于发现因果关系。在这项工作中，我们充分利用的先验知识框架，热启动因果关系发现。我们发现，Bucker-DAG提供了更大的灵活性比现有的方法的基础上分层的背景知识，并介绍了两个修改后的基于约束的算法，Bucker-PC和Bucker-FCI，因果关系的发现在完全和部分观察设置，分别。对模拟数据的实证评估表明，在没有先验知识的情况下，PC和FCI都优于各自的基线。
摘要：Finding cause-effect relationships is of key importance in science. Causal discovery aims to recover a graph from data that succinctly describes these cause-effect relationships. However, current methods face several challenges, especially when dealing with high-dimensional data and complex dependencies. Incorporating prior knowledge about the system can aid causal discovery. In this work, we leverage Cluster-DAGs as a prior knowledge framework to warm-start causal discovery. We show that Cluster-DAGs offer greater flexibility than existing approaches based on tiered background knowledge and introduce two modified constraint-based algorithms, Cluster-PC and Cluster-FCI, for causal discovery in the fully and partially observed setting, respectively. Empirical evaluation on simulated data demonstrates that Cluster-PC and Cluster-FCI outperform their respective baselines without prior knowledge.
联邦学习|隐私保护|加密(1篇)
【1】HybridVFL: Disentangled Feature Learning for Edge-Enabled Vertical Federated Multimodal Classification
标题：HybridVFL：边缘启用垂直联邦多模式分类的分离特征学习
链接：https://arxiv.org/abs/2512.10701
备注：6 pages, 2 figures, 1 table. Accepted at UCC '25 (IEEE/ACM 18th International Conference on Utility and Cloud Computing), December 1-4, 2025, Nantes, France. DOI to be activated upon final publication
摘要：垂直联合学习（VFL）为边缘AI场景提供了一种隐私保护范例，如移动健康诊断，其中敏感的多模态数据驻留在分布式，资源受限的设备上。然而，标准的VFL系统往往遭受性能限制，由于过于简单的功能融合。本文介绍了HybridVFL，一种新的框架，旨在克服这一瓶颈，采用客户端功能解开配对与服务器端的跨模态Transformer上下文感知融合。通过对多模态HAM10000皮肤病变数据集的系统评估，我们证明了HybridVFL的性能显著优于标准联邦基线，验证了高级融合机制在强大的隐私保护系统中的重要性。
摘要：Vertical Federated Learning (VFL) offers a privacy-preserving paradigm for Edge AI scenarios like mobile health diagnostics, where sensitive multimodal data reside on distributed, resource-constrained devices. Yet, standard VFL systems often suffer performance limitations due to simplistic feature fusion. This paper introduces HybridVFL, a novel framework designed to overcome this bottleneck by employing client-side feature disentanglement paired with a server-side cross-modal transformer for context-aware fusion. Through systematic evaluation on the multimodal HAM10000 skin lesion dataset, we demonstrate that HybridVFL significantly outperforms standard federated baselines, validating the criticality of advanced fusion mechanisms in robust, privacy-preserving systems.
推理|分析|理解|解释(4篇)
【1】Causal Reasoning Favors Encoders: On The Limits of Decoder-Only Models
标题：因果推理有利于编码器：论仅解码器模型的局限性
链接：https://arxiv.org/abs/2512.10561
摘要：在上下文学习（ICL）支持大型语言模型（LLM）的最新进展，尽管它在因果推理中的作用和表现仍不清楚。因果推理需要多跳合成和严格的连接控制，依赖于输入的虚假词汇关系可能会产生误导性的结果。我们假设，由于他们的能力，项目的输入到一个潜在的空间，编码器和编码器解码器架构更适合于多跳合取推理与解码器只模型。为此，我们比较了自然语言和非自然语言场景中所有上述体系结构的微调版本，以及零和Few Shot ICL。我们发现，ICL本身是不足以可靠的因果推理，往往过分关注不相关的输入功能。特别是，仅解码器模型对分布偏移明显脆弱，而微调的编码器和编码器解码器模型可以在我们的测试中更稳健地推广，包括非自然语言分裂。这两种架构仅在大规模上被仅解码器架构匹配或超越。我们的结论是，注意到成本效益，短时间内强大的因果推理，编码器或编码器解码器架构有针对性的微调是优选的。
摘要：In context learning (ICL) underpins recent advances in large language models (LLMs), although its role and performance in causal reasoning remains unclear. Causal reasoning demands multihop composition and strict conjunctive control, and reliance on spurious lexical relations of the input could provide misleading results. We hypothesize that, due to their ability to project the input into a latent space, encoder and encoder decoder architectures are better suited for said multihop conjunctive reasoning versus decoder only models. To do this, we compare fine-tuned versions of all the aforementioned architectures with zero and few shot ICL in both natural language and non natural language scenarios. We find that ICL alone is insufficient for reliable causal reasoning, often overfocusing on irrelevant input features. In particular, decoder only models are noticeably brittle to distributional shifts, while finetuned encoder and encoder decoder models can generalize more robustly across our tests, including the non natural language split. Both architectures are only matched or surpassed by decoder only architectures at large scales. We conclude by noting that for cost effective, short horizon robust causal reasoning, encoder or encoder decoder architectures with targeted finetuning are preferable.
【2】Disentangled and Distilled Encoder for Out-of-Distribution Reasoning with Rademacher Guarantees
标题：用于具有Rademacher保证的分布外推理的解开和蒸馏编码器
链接：https://arxiv.org/abs/2512.10522
摘要：最近，解纠缠的潜在空间的变分自动编码器（VAE）已被用来推理多标签的分布外（OOD）的测试样本，来自不同的分布比训练样本。解纠缠的潜在空间是指在潜在维度和图像的生成因素或重要特征之间具有一对多的映射。本文提出了一种解纠缠蒸馏编码器（DDE）框架，以减少部署在资源受限的设备上的OOD推理机的大小，同时保持解纠缠。DDE形式化的学生-教师蒸馏模型压缩作为一个约束优化问题，同时保持解纠缠与解纠缠约束。建立了基于Rademacher复杂性的蒸馏过程中解纠缠的理论保证。该方法是通过将压缩模型部署在NVIDIA上进行经验评估的
摘要：Recently, the disentangled latent space of a variational autoencoder (VAE) has been used to reason about multi-label out-of-distribution (OOD) test samples that are derived from different distributions than training samples. Disentangled latent space means having one-to-many maps between latent dimensions and generative factors or important characteristics of an image. This paper proposes a disentangled distilled encoder (DDE) framework to decrease the OOD reasoner size for deployment on resource-constrained devices while preserving disentanglement. DDE formalizes student-teacher distillation for model compression as a constrained optimization problem while preserving disentanglement with disentanglement constraints. Theoretical guarantees for disentanglement during distillation based on Rademacher complexity are established. The approach is evaluated empirically by deploying the compressed model on an NVIDIA
【3】Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
标题：具有稀疏自动编码器的可解释嵌入：数据分析工具包
链接：https://arxiv.org/abs/2512.10092
备注：Code: https://github.com/nickjiang2378/interp_embed
摘要：分析大规模文本语料库是机器学习的核心挑战，对于识别训练数据中的不良模型行为或偏差等任务至关重要。目前的方法通常依赖于昂贵的基于LLM的技术（例如，注释数据集差异）或密集嵌入模型（例如，用于聚类），这缺乏对感兴趣属性的控制。我们建议使用稀疏自动编码器（SAE）来创建SAE嵌入：其维度映射到可解释概念的表示。通过四个数据分析任务，我们表明，SAE嵌入比LLM更具成本效益和可靠性，比密集嵌入更可控。使用SAE的大假设空间，我们可以揭示诸如（1）数据集之间的语义差异和（2）文档中意外的概念相关性等见解。例如，通过比较模型响应，我们发现Grok-4比其他九个前沿模型更经常地澄清模糊性。相对于LLM，SAE嵌入以2- 8倍的低成本揭示了更大的差异，并更可靠地识别偏差。此外，SAE嵌入是可控的：通过过滤概念，我们可以（3）沿着感兴趣的轴聚类文档，（4）在基于属性的检索上优于密集嵌入。使用SAE嵌入，我们通过两个案例研究模型行为：调查OpenAI模型行为如何随时间变化，并找到Tulu-3学习的“触发”短语（Lambert et al.，2024年）的训练数据。这些结果将SAE定位为非结构化数据分析的通用工具，并突出了通过其数据解释模型的重要性。
摘要：Analyzing large-scale text corpora is a core challenge in machine learning, crucial for tasks like identifying undesirable model behaviors or biases in training data. Current methods often rely on costly LLM-based techniques (e.g. annotating dataset differences) or dense embedding models (e.g. for clustering), which lack control over the properties of interest. We propose using sparse autoencoders (SAEs) to create SAE embeddings: representations whose dimensions map to interpretable concepts. Through four data analysis tasks, we show that SAE embeddings are more cost-effective and reliable than LLMs and more controllable than dense embeddings. Using the large hypothesis space of SAEs, we can uncover insights such as (1) semantic differences between datasets and (2) unexpected concept correlations in documents. For instance, by comparing model responses, we find that Grok-4 clarifies ambiguities more often than nine other frontier models. Relative to LLMs, SAE embeddings uncover bigger differences at 2-8x lower cost and identify biases more reliably. Additionally, SAE embeddings are controllable: by filtering concepts, we can (3) cluster documents along axes of interest and (4) outperform dense embeddings on property-based retrieval. Using SAE embeddings, we study model behavior with two case studies: investigating how OpenAI model behavior has changed over time and finding "trigger" phrases learned by Tulu-3 (Lambert et al., 2024) from its training data. These results position SAEs as a versatile tool for unstructured data analysis and highlight the neglected importance of interpreting models through their data.
【4】Error Analysis of Generalized Langevin Equations with Approximated Memory Kernels
标题：具有逼近记忆核的广义朗之万方程的误差分析
链接：https://arxiv.org/abs/2512.10256
摘要：我们分析了具有记忆的随机动力系统的预测误差，重点是广义Langevin方程（GLEs）制定为随机Volterra方程。我们建立，在一个强凸的潜力，轨迹的差异衰减的记忆内核的衰减所确定的速率，并定量地限制在一个加权范数的内核的估计误差。我们的分析集成了同步噪声耦合与Volterra比较定理，包括次指数和指数内核类。对于一阶模型，我们在加权空间中使用预解式估计得到矩和扰动界。对于二阶模型与限制势，我们证明了收缩和稳定的内核扰动下使用hypocortive的Lyapunov型距离。该框架适应非平移不变内核和白噪声强迫，明确地将改进的内核估计与增强的轨迹预测联系起来。数值例子验证了这些理论研究结果。
摘要：We analyze prediction error in stochastic dynamical systems with memory, focusing on generalized Langevin equations (GLEs) formulated as stochastic Volterra equations. We establish that, under a strongly convex potential, trajectory discrepancies decay at a rate determined by the decay of the memory kernel and are quantitatively bounded by the estimation error of the kernel in a weighted norm. Our analysis integrates synchronized noise coupling with a Volterra comparison theorem, encompassing both subexponential and exponential kernel classes. For first-order models, we derive moment and perturbation bounds using resolvent estimates in weighted spaces. For second-order models with confining potentials, we prove contraction and stability under kernel perturbations using a hypocoercive Lyapunov-type distance. This framework accommodates non-translation-invariant kernels and white-noise forcing, explicitly linking improved kernel estimation to enhanced trajectory prediction. Numerical examples validate these theoretical findings.
检测相关(2篇)
【1】Enhancing Fake-News Detection with Node-Level Topological Features
标题：利用节点级布局特征增强假新闻检测
链接：https://arxiv.org/abs/2512.09974
摘要：近年来，错误信息和假新闻的泛滥对个人和社会构成了严重威胁，促使人们对自动检测方法进行了深入研究。以前的工作表明，整合内容，用户偏好和传播结构可以实现强大的性能，但将所有的图级表示学习完全留给GNN，隐藏任何显式的拓扑线索。为了缩小这一差距，我们引入了一个轻量级的增强：对于每个节点，我们附加了两个经典的图论度量，度中心性和局部聚类系数，其原始的BERT和配置文件嵌入，从而明确标记的枢纽和社区的角色。在UPFD Politifact子集中，这个简单的修改将宏F1从原始基线的0.7753提升到0.8344。我们的研究不仅证明了显式拓扑特征在假新闻检测中的实用价值，而且还为其他信息扩散任务中的融合图度量提供了一个可解释的，易于复制的模板。
摘要 ：In recent years, the proliferation of misinformation and fake news has posed serious threats to individuals and society, spurring intense research into automated detection methods. Previous work showed that integrating content, user preferences, and propagation structure achieves strong performance, but leaves all graph-level representation learning entirely to the GNN, hiding any explicit topological cues. To close this gap, we introduce a lightweight enhancement: for each node, we append two classical graph-theoretic metrics, degree centrality and local clustering coefficient, to its original BERT and profile embeddings, thus explicitly flagging the roles of hub and community. In the UPFD Politifact subset, this simple modification boosts macro F1 from 0.7753 to 0.8344 over the original baseline. Our study not only demonstrates the practical value of explicit topology features in fake-news detection but also provides an interpretable, easily reproducible template for fusing graph metrics in other information-diffusion tasks.
【2】Exploring Health Misinformation Detection with Multi-Agent Debate
标题：基于多Agent辩论的健康错误信息检测研究
链接：https://arxiv.org/abs/2512.09935
摘要：随着网上错误信息的激增，对与健康相关的声明进行事实核查变得越来越重要。有效的核查既需要检索高质量的证据，也需要严格的推理过程。在本文中，我们提出了一个健康错误信息检测的两阶段框架：一致性评分预测，然后是多智能体辩论。在第一阶段，我们采用大型语言模型（LLM）来独立评估检索到的文章，并计算反映整体证据立场的聚合一致性得分。当该分数指示共识不足-低于预定义阈值时-系统进行到第二阶段。多个代理人参与结构化的辩论，以综合相互矛盾的证据，并产生合理的判决与明确的理由。实验结果表明，我们的两阶段的方法相比，基线方法实现了卓越的性能，突出了复杂的验证任务的自动评分与协作推理相结合的价值。
摘要：Fact-checking health-related claims has become increasingly critical as misinformation proliferates online. Effective verification requires both the retrieval of high-quality evidence and rigorous reasoning processes. In this paper, we propose a two-stage framework for health misinformation detection: Agreement Score Prediction followed by Multi-Agent Debate. In the first stage, we employ large language models (LLMs) to independently evaluate retrieved articles and compute an aggregated agreement score that reflects the overall evidence stance. When this score indicates insufficient consensus-falling below a predefined threshold-the system proceeds to a second stage. Multiple agents engage in structured debate to synthesize conflicting evidence and generate well-reasoned verdicts with explicit justifications. Experimental results demonstrate that our two-stage approach achieves superior performance compared to baseline methods, highlighting the value of combining automated scoring with collaborative reasoning for complex verification tasks.
分类|识别(4篇)
【1】Classifier Reconstruction Through Counterfactual-Aware Wasserstein Prototypes
标题：通过反事实意识的沃瑟斯坦原型重建分类器
链接：https://arxiv.org/abs/2512.10878
备注：Accepted by Actionable Interpretability Workshop at ICML 2025
摘要：反事实解释通过识别实现所需模型预测所需的最小输入变化来提供可操作的见解。除了它们的可解释性优势之外，反事实还可以用于模型重建，其中代理模型被训练以复制目标模型的行为。在这项工作中，我们证明了模型重建可以显着提高认识，反事实，这通常是靠近决策边界，可以作为信息，但代表性较低的样本，这两个类。这在对标记数据的访问受限的设置中特别有益。我们提出了一种方法，将原始数据样本与反事实近似类原型使用Wasserstein重心，从而保持每个类的底层分布结构。这种方法提高了代理模型的质量，并缓解了决策边界偏移的问题，当反事实被天真地视为普通训练实例时，通常会出现这种问题。在多个数据集上的实验结果表明，该方法提高了代理模型和目标模型之间的保真度，验证了其有效性。
摘要：Counterfactual explanations provide actionable insights by identifying minimal input changes required to achieve a desired model prediction. Beyond their interpretability benefits, counterfactuals can also be leveraged for model reconstruction, where a surrogate model is trained to replicate the behavior of a target model. In this work, we demonstrate that model reconstruction can be significantly improved by recognizing that counterfactuals, which typically lie close to the decision boundary, can serve as informative though less representative samples for both classes. This is particularly beneficial in settings with limited access to labeled data. We propose a method that integrates original data samples with counterfactuals to approximate class prototypes using the Wasserstein barycenter, thereby preserving the underlying distributional structure of each class. This approach enhances the quality of the surrogate model and mitigates the issue of decision boundary shift, which commonly arises when counterfactuals are naively treated as ordinary training instances. Empirical results across multiple datasets show that our method improves fidelity between the surrogate and target models, validating its effectiveness.
【2】Agile Deliberation: Concept Deliberation for Subjective Visual Classification
标题：敏捷审议：主观视觉分类的概念审议
链接：https://arxiv.org/abs/2512.10821
摘要：从内容审核到内容策展，需要视觉概念的视觉分类器的应用正在迅速扩展。现有的人在回路方法通常假设用户从一个清晰、稳定的概念理解开始，以便能够提供高质量的监督。在现实中，用户通常从一个模糊的想法开始，必须通过“概念审议”迭代地完善它，这是我们通过与内容审核专家的结构化访谈发现的一种做法。我们将真实内容主持人在审议中使用的常见策略操作成一个名为“敏捷审议”的人在回路框架，该框架明确支持不断发展的主观概念。该系统通过让用户接触边界案例，支持用户自己定义概念。该系统通过两个审议阶段来实现这一点：（1）概念范围界定，其将初始概念分解成子概念的结构化层级，以及（2）概念迭代，其使语义上的边界示例表面化以供用户反映和反馈，从而迭代地将图像分类器与用户的演进意图对准。由于概念审议本质上是主观的和互动的，我们通过18个用户会话，每个1.5小时长，而不是标准的基准数据集，煞费苦心地评估框架。我们发现，敏捷审议实现了7.5%的F1分数比自动分解基线和超过3%的高于手动审议，而参与者报告更清晰的概念理解和更低的认知努力。
摘要：From content moderation to content curation, applications requiring vision classifiers for visual concepts are rapidly expanding. Existing human-in-the-loop approaches typically assume users begin with a clear, stable concept understanding to be able to provide high-quality supervision. In reality, users often start with a vague idea and must iteratively refine it through "concept deliberation", a practice we uncovered through structured interviews with content moderation experts. We operationalize the common strategies in deliberation used by real content moderators into a human-in-the-loop framework called "Agile Deliberation" that explicitly supports evolving and subjective concepts. The system supports users in defining the concept for themselves by exposing them to borderline cases. The system does this with two deliberation stages: (1) concept scoping, which decomposes the initial concept into a structured hierarchy of sub-concepts, and (2) concept iteration, which surfaces semantically borderline examples for user reflection and feedback to iteratively align an image classifier with the user's evolving intent. Since concept deliberation is inherently subjective and interactive, we painstakingly evaluate the framework through 18 user sessions, each 1.5h long, rather than standard benchmarking datasets. We find that Agile Deliberation achieves 7.5% higher F1 scores than automated decomposition baselines and more than 3% higher than manual deliberation, while participants reported clearer conceptual understanding and lower cognitive effort.
【3】Sequence-to-Image Transformation for Sequence Classification Using Rips Complex Construction and Chaos Game Representation
标题：使用Rips复体结构和混乱博弈表示的序列到图像转换用于序列分类
链接：https://arxiv.org/abs/2512.10141
摘要 ：用于分子序列分类的传统特征工程方法存在稀疏性问题和计算复杂性，而深度学习模型通常在表格生物数据上表现不佳。本文介绍了一种新的拓扑方法，将混沌博弈表示（CGR）与代数拓扑学中的Rips复形构造相结合，将分子序列转化为图像。我们的方法通过CGR将序列元素映射到二维坐标，计算成对距离，并构建RIP复合体来捕获局部结构和全局拓扑特征。我们提供了形式上的保证表示的唯一性，拓扑稳定性和信息保存。在抗癌肽数据集上的广泛实验表明，与基于向量的序列语言模型和现有的基于图像的方法相比，性能优越，在乳腺癌和肺癌数据集上分别达到86.8%和94.5%的准确率。拓扑表示保留了关键的序列信息，同时能够有效利用基于视觉的深度学习架构进行分子序列分析。
摘要：Traditional feature engineering approaches for molecular sequence classification suffer from sparsity issues and computational complexity, while deep learning models often underperform on tabular biological data. This paper introduces a novel topological approach that transforms molecular sequences into images by combining Chaos Game Representation (CGR) with Rips complex construction from algebraic topology. Our method maps sequence elements to 2D coordinates via CGR, computes pairwise distances, and constructs Rips complexes to capture both local structural and global topological features. We provide formal guarantees on representation uniqueness, topological stability, and information preservation. Extensive experiments on anticancer peptide datasets demonstrate superior performance over vector-based, sequence language models, and existing image-based methods, achieving 86.8\% and 94.5\% accuracy on breast and lung cancer datasets, respectively. The topological representation preserves critical sequence information while enabling effective utilization of vision-based deep learning architectures for molecular sequence analysis.
【4】LxCIM: a new rank-based binary classifier performance metric invariant to local exchange of classes
标题：LxTIM：一种新的基于排名的二进制分类器性能指标，不受本地类交换的影响
链接：https://arxiv.org/abs/2512.10053
备注：28 pages, 7 figures
摘要：二进制分类是机器学习中最古老，最流行和研究的问题之一。然而，用于评估模型性能的指标得到了相对较少的关注。受试者工作特征曲线下面积（AUROC）长期以来一直是模型比较的标准选择。尽管它的优点，AUROC并不总是理想的，特别是对于本地交换类（LxC）不变的问题，在这项工作中引入了一种新形式的度量不变性。为了解决这一限制，我们提出了LxCIM（LxC-invariant metric），它不仅是基于排名的，在局部交换类下不变，而且直观，逻辑一致，始终可计算，同时通过累积准确率-决策率曲线进行更详细的分析。此外，LxCIM与AUROC、准确度和准确度-决策率曲线下面积（AUDRC）有明确的理论联系。这些关系允许多种互补的解释：作为AUROC的对称形式，基于等级的准确性模拟，或AUDRC的更具代表性和更可解释的变体。最后，我们证明了LxCIM的直接适用性的双变量因果发现问题（表现出不变性本地交换类），并显示它如何解决公认的局限性，在这一领域使用的现有指标。所有代码和实现细节都可以在github.com/tiagobrogueira/Causal-Discovery-In-Exchangeable-Data上公开获得。
摘要：Binary classification is one of the oldest, most prevalent, and studied problems in machine learning. However, the metrics used to evaluate model performance have received comparatively little attention. The area under the receiver operating characteristic curve (AUROC) has long been a standard choice for model comparison. Despite its advantages, AUROC is not always ideal, particularly for problems that are invariant to local exchange of classes (LxC), a new form of metric invariance introduced in this work. To address this limitation, we propose LxCIM (LxC-invariant metric), which is not only rank-based and invariant under local exchange of classes, but also intuitive, logically consistent, and always computable, while enabling more detailed analysis through the cumulative accuracy-decision rate curve. Moreover, LxCIM exhibits clear theoretical connections to AUROC, accuracy, and the area under the accuracy-decision rate curve (AUDRC). These relationships allow for multiple complementary interpretations: as a symmetric form of AUROC, a rank-based analogue of accuracy, or a more representative and more interpretable variant of AUDRC. Finally, we demonstrate the direct applicability of LxCIM to the bivariate causal discovery problem (which exhibits invariance to local exchange of classes) and show how it addresses the acknowledged limitations of existing metrics used in this field. All code and implementation details are publicly available at github.com/tiagobrogueira/Causal-Discovery-In-Exchangeable-Data.
表征(3篇)
【1】What matters for Representation Alignment: Global Information or Spatial Structure?
标题：对于表示对齐来说什么重要：全球信息还是空间结构？
链接：https://arxiv.org/abs/2512.10794
备注：Project page: https://end2end-diffusion.github.io/irepa
摘要：表示对齐（REPA）通过将表示从强大的预训练视觉编码器中提取到中间扩散特征来指导生成训练。我们研究了一个基本问题：目标表示的哪个方面对生成很重要，它的\textit{global} \revision{semantic}信息（例如，通过ImageNet-1 K精度测量）或其空间结构（即补丁令牌之间的成对余弦相似性）？流行的智慧认为，更强的全局语义性能导致更好的生成作为目标表示。为了研究这一点，我们首先对27种不同的视觉编码器和不同的模型尺度进行了大规模的实证分析。结果是令人惊讶的;空间结构，而不是全局性能，驱动目标表示的生成性能。为了进一步研究这一点，我们介绍了两个简单的修改，特别强调了空间信息的传输。我们用一个简单的卷积层取代REPA中的标准MLP投影层，并为外部表示引入空间归一化层。令人惊讶的是，我们的简单方法（在$
摘要：Representation alignment (REPA) guides generative training by distilling representations from a strong, pretrained vision encoder to intermediate diffusion features. We investigate a fundamental question: what aspect of the target representation matters for generation, its \textit{global} \revision{semantic} information (e.g., measured by ImageNet-1K accuracy) or its spatial structure (i.e. pairwise cosine similarity between patch tokens)? Prevalent wisdom holds that stronger global semantic performance leads to better generation as a target representation. To study this, we first perform a large-scale empirical analysis across 27 different vision encoders and different model scales. The results are surprising; spatial structure, rather than global performance, drives the generation performance of a target representation. To further study this, we introduce two straightforward modifications, which specifically accentuate the transfer of \emph{spatial} information. We replace the standard MLP projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation. Surprisingly, our simple method (implemented in $
【2】Neuronal Attention Circuit (NAC) for Representation Learning
标题：用于表象学习的神经元注意力回路（RAC）
链接：https://arxiv.org/abs/2512.10282
备注：Paper for ICML2026
摘要：注意力改善了RNN的表示学习，但其离散性限制了连续时间（CT）建模。我们介绍了神经元注意回路（NAC），一种新的，生物学上合理的CT-注意机制，将注意力逻辑计算重新表述为线性一阶常微分方程的解决方案，该方程具有来自重新利用的非线性互连门。elegans}神经元电路策略（NCP）布线机制。NAC将密集投影替换为用于关键查询投影的稀疏传感器门，以及具有用于计算\textit{content-target}和\textit{learnable time-constant}门的两个头的稀疏骨干网络，从而实现高效的自适应动态。NAC支持三种注意力logit计算模式：（i）显式欧拉积分，（ii）精确的封闭形式的解决方案，和（iii）稳态近似。为了提高内存强度，我们实现了一个稀疏的Top-\n {K}成对连接方案，该方案选择性地管理键查询交互。我们提供严格的理论保证，包括状态稳定性，有界逼近误差和通用逼近。从经验上讲，我们在不同的领域实施了NAC，包括不规则时间序列分类、自动驾驶车辆的车道保持和工业自动化。我们观察到，NAC匹配或优于竞争的准确性基线，并在运行时间和内存效率与几个CT基线相比，占据了中间位置。
摘要 ：Attention improves representation learning over RNNs, but its discrete nature limits continuous-time (CT) modeling. We introduce Neuronal Attention Circuit (NAC), a novel, biologically plausible CT-Attention mechanism that reformulates attention logits computation as the solution to a linear first-order ODE with nonlinear interlinked gates derived from repurposing \textit{C. elegans} Neuronal Circuit Policies (NCPs) wiring mechanism. NAC replaces dense projections with sparse sensory gates for key-query projections and a sparse backbone network with two heads for computing \textit{content-target} and \textit{learnable time-constant} gates, enabling efficient adaptive dynamics. NAC supports three attention logit computation modes: (i) explicit Euler integration, (ii) exact closed-form solution, and (iii) steady-state approximation. To improve memory intensity, we implemented a sparse Top-\emph{K} pairwise concatenation scheme that selectively curates key-query interactions. We provide rigorous theoretical guarantees, including state stability, bounded approximation errors, and universal approximation. Empirically, we implemented NAC in diverse domains, including irregular time-series classification, lane-keeping for autonomous vehicles, and industrial prognostics. We observed that NAC matches or outperforms competing baselines in accuracy and occupies an intermediate position in runtime and memory efficiency compared with several CT baselines.
【3】CHyLL: Learning Continuous Neural Representations of Hybrid Systems
标题：CHyLL：学习混合系统的连续神经表示
链接：https://arxiv.org/abs/2512.10117
摘要：学习具有连续和离散时间动态的混合系统的流是具有挑战性的。现有的方法学习每个离散模式中的动态，其遭受流中的模式切换和不连续性的组合。在这项工作中，我们提出了CHyLL（Continuous Hybrid System Learning in Latent Space），它可以学习混合系统的连续神经表示，而无需轨迹分割，事件函数或模式切换。CHyLL的关键见解是重置映射将状态空间粘合在保护表面上，将状态空间重新表示为分段光滑的商流形，其中流在空间上变得连续。基于这些见解和基于微分拓扑的嵌入定理，CHyLL同时学习高维空间中的无奇异性神经嵌入和连续流。我们展示了CHyLL可以准确地预测混合系统的流，并以优异的精度识别混合系统的拓扑不变量。最后，我们将CHyLL应用于随机最优控制问题。
摘要：Learning the flows of hybrid systems that have both continuous and discrete time dynamics is challenging. The existing method learns the dynamics in each discrete mode, which suffers from the combination of mode switching and discontinuities in the flows. In this work, we propose CHyLL (Continuous Hybrid System Learning in Latent Space), which learns a continuous neural representation of a hybrid system without trajectory segmentation, event functions, or mode switching. The key insight of CHyLL is that the reset map glues the state space at the guard surface, reformulating the state space as a piecewise smooth quotient manifold where the flow becomes spatially continuous. Building upon these insights and the embedding theorems grounded in differential topology, CHyLL concurrently learns a singularity-free neural embedding in a higher-dimensional space and the continuous flow in it. We showcase that CHyLL can accurately predict the flow of hybrid systems with superior accuracy and identify the topological invariants of the hybrid systems. Finally, we apply CHyLL to the stochastic optimal control problem.
编码器(1篇)
【1】Extrapolation of Periodic Functions Using Binary Encoding of Continuous Numerical Values
标题：利用连续数值的二进制编码进行周期函数的外推
链接：https://arxiv.org/abs/2512.10817
备注：Submitted to JMLR, under review
摘要：我们报告的发现，二进制编码允许神经网络推断超出其训练范围的周期函数。我们引入归一化基-2编码（NB 2 E）作为一种编码连续数值的方法，并证明，使用这种输入编码，香草多层感知器（MLP）成功地外推不同的周期信号，而无需事先知道它们的函数形式。内部激活分析表明，NB 2 E诱导位相位表示，使MLP学习和外推信号结构独立的位置。
摘要：We report the discovery that binary encoding allows neural networks to extrapolate periodic functions beyond their training bounds. We introduce Normalized Base-2 Encoding (NB2E) as a method for encoding continuous numerical values and demonstrate that, using this input encoding, vanilla multi-layer perceptrons (MLP) successfully extrapolate diverse periodic signals without prior knowledge of their functional form. Internal activation analysis reveals that NB2E induces bit-phase representations, enabling MLPs to learn and extrapolate signal structure independently of position.
优化|敛散性(7篇)
【1】Optimal transport unlocks end-to-end learning for single-molecule localization
标题：最佳传输解锁单分子定位的端到端学习
链接：https://arxiv.org/abs/2512.10683
摘要：单分子定位显微镜（SMLM）允许通过检测和定位单个荧光团（染色到观察到的标本上的荧光分子）来重建超出衍射极限的生物学相关结构，以重建超分辨图像。目前，高效的SMLM需要不重叠的发射荧光团，导致长的采集时间，这阻碍了活细胞成像。最近的深度学习方法可以处理更密集的发射，但它们依赖于非最大抑制（NMS）层的变体，不幸的是，这些层是不可微的，并且可能会通过其局部融合策略丢弃真阳性。在本演示中，我们将SMLM训练目标重新定义为一个集合匹配问题，推导出一个最优传输损失，从而消除了在推理过程中对NMS的需求，并实现了端到端训练。此外，我们提出了一个迭代神经网络，集成了我们的模型内的显微镜的光学系统的知识。合成基准和真实生物数据的实验表明，我们的新的损失函数和架构超过了最先进的中等和高发射密度。代码可在https://github.com/RSLLES/SHOT上获得。
摘要：Single-molecule localization microscopy (SMLM) allows reconstructing biology-relevant structures beyond the diffraction limit by detecting and localizing individual fluorophores -- fluorescent molecules stained onto the observed specimen -- over time to reconstruct super-resolved images. Currently, efficient SMLM requires non-overlapping emitting fluorophores, leading to long acquisition times that hinders live-cell imaging. Recent deep-learning approaches can handle denser emissions, but they rely on variants of non-maximum suppression (NMS) layers, which are unfortunately non-differentiable and may discard true positives with their local fusion strategy. In this presentation, we reformulate the SMLM training objective as a set-matching problem, deriving an optimal-transport loss that eliminates the need for NMS during inference and enables end-to-end training. Additionally, we propose an iterative neural network that integrates knowledge of the microscope's optical system inside our model. Experiments on synthetic benchmarks and real biological data show that both our new loss function and architecture surpass the state of the art at moderate and high emitter densities. Code is available at https://github.com/RSLLES/SHOT.
【2】Multi-Objective Reward and Preference Optimization: Theory and Algorithms
标题：多目标奖励和偏好优化：理论与算法
链接：https://arxiv.org/abs/2512.10601
备注：PhD thesis
摘要：本论文开发了理论框架和算法，推进约束强化学习（RL）跨控制，偏好学习和大型语言模型的对齐。第一个贡献地址约束马尔可夫决策过程（CMDPs）的平均成本标准下，通过平均约束策略优化（ACPO）算法。ACPO将灵敏度分析与信赖域更新相结合，以确保稳定的约束处理，实现最先进的经验性能和理论保证。约束RL，然后扩展到有限的地平线设置通过e-COP，第一个政策优化方法的情节CMDPs。建立在一个情节的政策差异引理，e-COP提供可证明的性能，简单性和可扩展性，在安全关键的环境。然后，本文研究了从人类偏好的强化学习。warmPref-PS为线性强盗引入了一种后验抽样策略，该策略将来自异构评分员的离线偏好数据集成到在线学习中。明确建模的评分能力产生大量的遗憾减少和更有效的数据收集RLHF。PSPL算法通过从成对轨迹比较中联合采样奖励模型和过渡动态来进一步推进基于偏好的强化学习，提供贝叶斯简单遗憾保证和最优策略的鲁棒经验识别。最后的贡献适用于大规模的模型对齐这些方法。多目标约束优化视图产生MOPO，这是一种具有封闭形式更新的迭代算法，可扩展到数十亿参数的语言模型，并在对齐设置中保持稳健。总的来说，论文统一了平均成本，情节和偏好驱动的范式之间的约束强化学习，为安全和一致的决策提供理论进步和实用工具。
摘要：This thesis develops theoretical frameworks and algorithms that advance constrained reinforcement learning (RL) across control, preference learning, and alignment of large language models. The first contribution addresses constrained Markov Decision Processes (CMDPs) under the average-cost criterion through the Average-Constrained Policy Optimization (ACPO) algorithm. ACPO integrates sensitivity analysis with trust-region updates to ensure stable constraint handling, achieving state-of-the-art empirical performance with theoretical guarantees. Constrained RL is then extended to finite-horizon settings via e-COP, the first policy optimization method for episodic CMDPs. Built on an episodic policy difference lemma, e-COP offers provable performance, simplicity, and scalability in safety-critical environments. The thesis then investigates reinforcement learning from human preferences. warmPref-PS introduces a posterior sampling strategy for linear bandits that integrates offline preference data from heterogeneous raters into online learning. Explicit modeling of rater competence yields substantial regret reduction and more efficient data collection for RLHF. The PSPL algorithm further advances preference-based RL by jointly sampling reward models and transition dynamics from pairwise trajectory comparisons, providing Bayesian simple-regret guarantees and robust empirical identification of optimal policies. The final contribution applies these methods to large-scale model alignment. A multi-objective constrained optimization view yields MOPO, an iterative algorithm with closed-form updates that scales to multi-billion-parameter language models and remains robust across alignment settings. Collectively, the thesis unifies constrained RL across average-cost, episodic, and preference-driven paradigms, delivering theoretical advances and practical tools for safe and aligned decision-making.
【3】Hybrid Learning and Optimization-Based Dynamic Scheduling for DL Workloads on Heterogeneous GPU Clusters
标题：异构GPU集群上基于混合学习和优化的DL负载动态调度
链接：https://arxiv.org/abs/2512.10271
摘要：现代云平台越来越多地承载大规模深度学习（DL）工作负载，需要高吞吐量、低延迟的GPU调度。然而，GPU集群日益增长的异构性和对应用程序特性的有限可见性对现有的并行计算器构成了重大挑战，这些并行计算器通常依赖于离线分析或特定于应用程序的假设。我们提出了RLTune，一个基于应用程序不可知的强化学习（RL）的调度框架，动态优先级和分配异构GPU集群上的DL作业。RLTune集成了RL驱动的优先级与基于MILP的作业到节点映射，以优化系统范围的目标，如作业完成时间（JCT），排队延迟和资源利用率。经过Microsoft Philly、Helios和阿里巴巴的大规模生产跟踪培训，RLTune将GPU利用率提高了20%，将延迟减少了81%，并将JCT缩短了70%。与以前的方法不同，RLTune可以在不同的工作负载之间进行推广，而不需要进行每个作业的分析，这使得云提供商可以大规模部署，以实现更高效、公平和可持续的DL工作负载管理。
摘要：Modern cloud platforms increasingly host large-scale deep learning (DL) workloads, demanding high-throughput, low-latency GPU scheduling. However, the growing heterogeneity of GPU clusters and limited visibility into application characteristics pose major challenges for existing schedulers, which often rely on offline profiling or application-specific assumptions. We present RLTune, an application-agnostic reinforcement learning (RL)-based scheduling framework that dynamically prioritizes and allocates DL jobs on heterogeneous GPU clusters. RLTune integrates RL-driven prioritization with MILP-based job-to-node mapping to optimize system-wide objectives such as job completion time (JCT), queueing delay, and resource utilization. Trained on large-scale production traces from Microsoft Philly, Helios, and Alibaba, RLTune improves GPU utilization by up to 20%, reduces queueing delay by up to 81%, and shortens JCT by as much as 70 percent. Unlike prior approaches, RLTune generalizes across diverse workloads without requiring per-job profiling, making it practical for cloud providers to deploy at scale for more efficient, fair, and sustainable DL workload management.
【4】Distributionally Robust Regret Optimal Control Under Moment-Based Ambiguity Sets
标题：基于动量的模糊集下的分布鲁棒遗憾最优控制
链接：https://arxiv.org/abs/2512.10906
备注：21 pages, 2 figures
摘要：在本文中，我们考虑一类有限时域，线性二次随机控制问题，其中的概率分布的噪声过程是未知的，但假设属于一个模糊集，由所有的分布，其均值和协方差位于以给定的标称值为中心的范数球。为了解决分布模糊性，我们探索因果仿射控制策略的设计，以最大限度地减少最坏情况下的预期遗憾在给定的模糊性集的所有分布。由此产生的极大极小最优控制问题，承认一个等价的重新作为一个听话的凸规划，对应于正规化版本的名义线性二次随机控制问题。虽然这个凸规划可以被改写为半定规划，但半定规划通常使用原始-对偶内点方法来求解，这种方法在实践中对问题的规模影响很小。为了解决这个问题，我们提出了一个可扩展的双重投影次梯度方法来计算最优控制器的任意精度。数值实验基准所提出的方法对国家的最先进的数据驱动和分布式鲁棒控制设计方法。
摘要：In this paper, we consider a class of finite-horizon, linear-quadratic stochastic control problems, where the probability distribution governing the noise process is unknown but assumed to belong to an ambiguity set consisting of all distributions whose mean and covariance lie within norm balls centered at given nominal values. To address the distributional ambiguity, we explore the design of causal affine control policies to minimize the worst-case expected regret over all distributions in the given ambiguity set. The resulting minimax optimal control problem is shown to admit an equivalent reformulation as a tractable convex program that corresponds to a regularized version of the nominal linear-quadratic stochastic control problem. While this convex program can be recast as a semidefinite program, semidefinite programs are typically solved using primal-dual interior point methods that scale poorly with the problem size in practice. To address this limitation, we propose a scalable dual projected subgradient method to compute optimal controllers to an arbitrary accuracy. Numerical experiments are presented to benchmark the proposed method against state-of-the-art data-driven and distributionally robust control design approaches.
【5】Physics-informed Polynomial Chaos Expansion with Enhanced Constrained Optimization Solver and D-optimal Sampling
标题：利用增强型约束优化求解器和D-最优采样进行物理信息的多项混乱扩展
链接：https://arxiv.org/abs/2512.10873
摘要：物理多项式混沌展开（PC$^2$）通过将控制方程和其他物理约束嵌入到标准数据驱动的多项式混沌展开（PCE）中并通过Karush-Kuhn-Tucker（KKT）条件求解，提供了一种有效的物理约束代理建模框架。该方法提高了代理模型的物理可解释性，同时实现了高计算效率和准确性。然而，性能和效率的PC $^2 $仍然可以降低高维参数空间，有限的数据可用性，或不具有代表性的训练数据。为了解决这个问题，本研究探讨了PC $^2 $框架的两个补充增强。首先，一个数值上有效的约束优化求解器，直接更新的拉格朗日乘子（SULM），采用作为替代传统的KKT求解器。SULM方法显着降低了计算成本时，解决物理约束的问题，高维和衍生的边界条件，需要大量的虚拟点。其次，采用D-最优采样策略选择信息量大的虚拟点，以提高算法的稳定性，实现算法精度和效率的平衡。所提出的方法被集成到PC的框架和评估通过数值例子的代表性的物理系统由常微分方程或偏微分方程。结果表明，改进后的PC$^2$比标准PC$^2$具有更好的综合能力，非常适合于高维不确定性量化任务.
摘要 ：Physics-informed polynomial chaos expansions (PC$^2$) provide an efficient physically constrained surrogate modeling framework by embedding governing equations and other physical constraints into the standard data-driven polynomial chaos expansions (PCE) and solving via the Karush-Kuhn-Tucker (KKT) conditions. This approach improves the physical interpretability of surrogate models while achieving high computational efficiency and accuracy. However, the performance and efficiency of PC$^2$ can still be degraded with high-dimensional parameter spaces, limited data availability, or unrepresentative training data. To address this problem, this study explores two complementary enhancements to the PC$^2$ framework. First, a numerically efficient constrained optimization solver, straightforward updating of Lagrange multipliers (SULM), is adopted as an alternative to the conventional KKT solver. The SULM method significantly reduces computational cost when solving physically constrained problems with high-dimensionality and derivative boundary conditions that require a large number of virtual points. Second, a D-optimal sampling strategy is utilized to select informative virtual points to improve the stability and achieve the balance of accuracy and efficiency of the PC$^2$. The proposed methods are integrated into the PC$^2$ framework and evaluated through numerical examples of representative physical systems governed by ordinary or partial differential equations. The results demonstrate that the enhanced PC$^2$ has better comprehensive capability than standard PC$^2$, and is well-suited for high-dimensional uncertainty quantification tasks.
【6】An Elementary Proof of the Near Optimality of LogSumExp Smoothing
标题：LogSumExp平滑近乎最优性的初步证明
链接：https://arxiv.org/abs/2512.10825
备注：10 pages
摘要：我们考虑在无穷范数下，在$\mathbb{R}^d$中的（坐标方向的）max函数的光滑化的设计。LogSumExp函数$f（x）=\ln（\sum^d_i\exp（x_i））$提供了一个经典的平滑，与max函数的值最多相差$\ln（d）$。我们提供了一个基本的建设的下限，建立每个高估平滑的最大值函数必须相差至少$\sim 0.8145\ln（d）$。因此，LogSumExp对于常数因子是最优的。然而，在小尺寸，我们提供更强大的，完全最优的平滑达到我们的下限，表明基于熵的LogSumExp平滑方法并不是最优的。
摘要：We consider the design of smoothings of the (coordinate-wise) max function in $\mathbb{R}^d$ in the infinity norm. The LogSumExp function $f(x)=\ln(\sum^d_i\exp(x_i))$ provides a classical smoothing, differing from the max function in value by at most $\ln(d)$. We provide an elementary construction of a lower bound, establishing that every overestimating smoothing of the max function must differ by at least $\sim 0.8145\ln(d)$. Hence, LogSumExp is optimal up to constant factors. However, in small dimensions, we provide stronger, exactly optimal smoothings attaining our lower bound, showing that the entropy-based LogSumExp approach to smoothing is not exactly optimal.
【7】The Interplay of Statistics and Noisy Optimization: Learning Linear Predictors with Random Data Weights
标题：统计和噪音优化的相互作用：使用随机数据权重学习线性预测器
链接：https://arxiv.org/abs/2512.10188
摘要：我们分析梯度下降与随机加权的数据点在线性回归模型，在一个通用的加权分布。这包括各种形式的随机梯度下降、重要性采样，但也扩展到具有任意连续值的加权分布，从而提供了一个统一的框架来分析各种噪声对训练轨迹的影响。我们刻画了通过随机加权引起的隐式正则化，将其与加权线性回归联系起来，并推导出一阶和二阶矩收敛的非渐近界。利用几何矩收缩，我们还研究了由附加噪声引起的平稳分布。基于这些结果，我们讨论了加权分布的具体选择如何影响潜在的优化问题和所得估计的统计特性，以及一些例子，导致快速收敛的权重导致统计性能不佳。
摘要：We analyze gradient descent with randomly weighted data points in a linear regression model, under a generic weighting distribution. This includes various forms of stochastic gradient descent, importance sampling, but also extends to weighting distributions with arbitrary continuous values, thereby providing a unified framework to analyze the impact of various kinds of noise on the training trajectory. We characterize the implicit regularization induced through the random weighting, connect it with weighted linear regression, and derive non-asymptotic bounds for convergence in first and second moments. Leveraging geometric moment contraction, we also investigate the stationary distribution induced by the added noise. Based on these results, we discuss how specific choices of weighting distribution influence both the underlying optimization problem and statistical properties of the resulting estimator, as well as some examples for which weightings that lead to fast convergence cause bad statistical performance.
预测|估计(9篇)
【1】A Kernel-based Resource-efficient Neural Surrogate for Multi-fidelity Prediction of Aerodynamic Field
标题：基于核的资源高效的气动力场多保真预测神经代理
链接：https://arxiv.org/abs/2512.10287
备注：24 pages, 15 figures
摘要：替代模型为昂贵的空气动力学模拟提供了快速的替代方案，在设计和优化应用中非常有用。这项研究提出了使用最近的基于内核的神经代理，KHRONOS。在这项工作中，我们混合稀疏的高保真度（HF）数据与低保真度（LF）的信息来预测不同的约束条件下的计算资源的空气动力场。与传统方法不同，KHRONOS是建立在变分原理，插值理论和张量分解。与密集神经网络相比，这些元素为大量修剪提供了数学基础。使用AirfRANS数据集作为高保真度基准，使用NeuralFoil生成低保真度对应物，这项工作将KHRONOS的性能与三种当代模型架构进行了比较：多层感知器（MLP），图形神经网络（GNN）和物理信息神经网络（PINN）。我们考虑不同级别的高保真数据可用性（0%，10%和30%）和日益复杂的几何参数化。这些用于预测翼型上的表面压力系数分布。结果表明，虽然所有模型最终都达到了相当的预测精度，但KHRONOS在资源受限的条件下表现出色。在这个领域，KHRONOS始终需要数量级更少的可训练参数，并以可比的精度提供比当代密集神经网络更快的训练和推理。这些发现突出了KHRONOS和类似架构在多保真度空气动力场预测中平衡精度和效率的潜力。
摘要：Surrogate models provide fast alternatives to costly aerodynamic simulations and are extremely useful in design and optimization applications. This study proposes the use of a recent kernel-based neural surrogate, KHRONOS. In this work, we blend sparse high-fidelity (HF) data with low-fidelity (LF) information to predict aerodynamic fields under varying constraints in computational resources. Unlike traditional approaches, KHRONOS is built upon variational principles, interpolation theory, and tensor decomposition. These elements provide a mathematical basis for heavy pruning compared to dense neural networks. Using the AirfRANS dataset as a high-fidelity benchmark and NeuralFoil to generate low-fidelity counterparts, this work compares the performance of KHRONOS with three contemporary model architectures: a multilayer perceptron (MLP), a graph neural network (GNN), and a physics-informed neural network (PINN). We consider varying levels of high-fidelity data availability (0%, 10%, and 30%) and increasingly complex geometry parameterizations. These are used to predict the surface pressure coefficient distribution over the airfoil. Results indicate that, whilst all models eventually achieve comparable predictive accuracy, KHRONOS excels in resource-constrained conditions. In this domain, KHRONOS consistently requires orders of magnitude fewer trainable parameters and delivers much faster training and inference than contemporary dense neural networks at comparable accuracy. These findings highlight the potential of KHRONOS and similar architectures to balance accuracy and efficiency in multi-fidelity aerodynamic field prediction.
【2】Partitioning the Sample Space for a More Precise Shannon Entropy Estimation
标题：基于样本空间划分的Shannon熵估计
链接：https://arxiv.org/abs/2512.10133
备注 ：The manuscript contains 6 pages and 10 figures. It has been accepted for International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA 2026)
摘要：可靠的数据驱动估计香农熵从小数据集，其中的例子的数量可能小于可能的结果的数量，是一个关键问题，在几个应用程序。在本文中，我们引入了一个离散熵估计，在这里我们使用的可分解属性结合估计的丢失的质量和数量的看不见的结果，以补偿由它们引起的负偏差。实验结果表明，该方法优于一些经典的估计在欠采样制度，并执行与一些公认的国家的最先进的估计。
摘要：Reliable data-driven estimation of Shannon entropy from small data sets, where the number of examples is potentially smaller than the number of possible outcomes, is a critical matter in several applications. In this paper, we introduce a discrete entropy estimator, where we use the decomposability property in combination with estimations of the missing mass and the number of unseen outcomes to compensate for the negative bias induced by them. Experimental results show that the proposed method outperforms some classical estimators in undersampled regimes, and performs comparably with some well-established state-of-the-art estimators.
【3】Independent Density Estimation
标题：独立密度估计
链接：https://arxiv.org/abs/2512.10067
备注：10 pages, 1 table, 4 figures
摘要：大规模视觉语言模型在图像字幕和条件图像生成等领域取得了显著的成果。尽管如此，这些模型在实现类似人类的合成概括方面仍然遇到困难。在这项研究中，我们提出了一种称为独立密度估计（IDE）的新方法来应对这一挑战。IDE旨在学习句子中的单个单词与图像中相应特征之间的联系，从而实现合成概括。我们建立了两个模型的基础上的IDE的哲学。第一个利用完全解开的视觉表示作为输入，第二个利用变分自动编码器从原始图像中获得部分解开的特征。此外，我们提出了一个基于熵的组合推理方法，结合预测的每个词在句子中。与当前模型相比，我们的模型在各种数据集上进行评估时，表现出对看不见的成分的卓越概括。
摘要：Large-scale Vision-Language models have achieved remarkable results in various domains, such as image captioning and conditioned image generation. Neverthe- less, these models still encounter difficulties in achieving human-like composi- tional generalization. In this study, we propose a new method called Independent Density Estimation (IDE) to tackle this challenge. IDE aims to learn the connec- tion between individual words in a sentence and the corresponding features in an image, enabling compositional generalization. We build two models based on the philosophy of IDE. The first one utilizes fully disentangled visual representations as input, and the second leverages a Variational Auto-Encoder to obtain partially disentangled features from raw images. Additionally, we propose an entropy- based compositional inference method to combine predictions of each word in the sentence. Our models exhibit superior generalization to unseen compositions compared to current models when evaluated on various datasets.
【4】Mitigating Exposure Bias in Risk-Aware Time Series Forecasting with Soft Tokens
标题：使用软令牌缓解风险意识时间序列预测中的暴露偏差
链接：https://arxiv.org/abs/2512.10056
摘要：自回归预测是糖尿病和血流动力学管理中预测控制的核心，其中不同的操作区域具有不同的临床风险。用教师强迫训练的标准模型受到暴露偏差的影响，为闭环使用产生不稳定的多步预测。我们引入了软令牌轨迹预测（SoTra），它传播连续概率分布（“软令牌”），以减轻暴露偏差并学习校准的不确定性感知轨迹。然后，风险感知解码模块可以最大限度地减少预期的临床伤害。在血糖预测中，SoTra将平均区域风险降低了18%;在血压预测中，它将有效临床风险降低了约15%。这些改进支持其在安全关键预测控制中的使用。
摘要：Autoregressive forecasting is central to predictive control in diabetes and hemodynamic management, where different operating zones carry different clinical risks. Standard models trained with teacher forcing suffer from exposure bias, yielding unstable multi-step forecasts for closed-loop use. We introduce Soft-Token Trajectory Forecasting (SoTra), which propagates continuous probability distributions (``soft tokens'') to mitigate exposure bias and learn calibrated, uncertainty-aware trajectories. A risk-aware decoding module then minimizes expected clinical harm. In glucose forecasting, SoTra reduces average zone-based risk by 18\%; in blood-pressure forecasting, it lowers effective clinical risk by approximately 15\%. These improvements support its use in safety-critical predictive control.
【5】DB2-TransF: All You Need Is Learnable Daubechies Wavelets for Time Series Forecasting
标题：DB 2-TransF：您需要的只是学习Daubechies波浪式时间序列预测
链接：https://arxiv.org/abs/2512.10051
摘要：时间序列预测需要能够有效捕获复杂时间依赖关系的模型，特别是在大规模和高维环境中。虽然基于Transformer的架构擅长于对远程依赖关系进行建模，但其二次计算复杂性对可扩展性和适应性造成了限制。为了克服这些挑战，我们引入了DB2-TransF，这是一种新颖的受Transformer启发的架构，它用可学习的Daubechies小波系数层取代了自注意机制。这个基于小波的模块有效地捕获多尺度局部和全局模式，并增强了时间序列预测任务中多个时间序列之间的相关性建模。在13个标准预测基准上进行的大量实验表明，DB2-TransF实现了与传统Transformers相当或更高的预测精度，同时大大减少了时间序列预测任务的内存使用。所获得的实验结果将DB2-TransF定位为用于高级时间序列预测的可扩展且资源高效的框架。我们的代码可在https://github.com/SteadySurfdom/DB2-TransF上获得
摘要：Time series forecasting requires models that can efficiently capture complex temporal dependencies, especially in large-scale and high-dimensional settings. While Transformer-based architectures excel at modeling long-range dependencies, their quadratic computational complexity poses limitations on scalability and adaptability. To overcome these challenges, we introduce DB2-TransF, a novel Transformer-inspired architecture that replaces the self-attention mechanism with a learnable Daubechies wavelet coefficient layer. This wavelet-based module efficiently captures multi-scale local and global patterns and enhances the modeling of correlations across multiple time series for the time series forecasting task. Extensive experiments on 13 standard forecasting benchmarks demonstrate that DB2-TransF achieves comparable or superior predictive accuracy to conventional Transformers, while substantially reducing memory usage for the time series forecasting task. The obtained experimental results position DB2-TransF as a scalable and resource-efficient framework for advanced time series forecasting. Our code is available at https://github.com/SteadySurfdom/DB2-TransF
【6】SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation
标题：SEMDICE：通过平稳分布修正估计实现非政策状态熵最大化
链接：https://arxiv.org/abs/2512.10042
备注：ICLR 2025
摘要：在强化学习的无监督预训练中，智能体旨在学习下游任务的先验策略，而不依赖于特定于任务的奖励函数。我们专注于状态熵最大化（SEM），其目标是学习最大化状态平稳分布的熵的策略。在本文中，我们介绍了SEMDICE，一个原则性的关闭政策算法，计算一个SEM政策从任意关闭政策数据集，直接优化的政策在空间内的平稳分布。SEMDICE计算一个单一的，稳定的马尔可夫状态熵最大化的政策，从一个任意的政策数据集。实验结果表明，SEMDICE在最大化状态熵方面优于基线算法，同时在基于SEM的无监督RL预训练方法中实现下游任务的最佳适应效率。
摘要 ：In the unsupervised pre-training for reinforcement learning, the agent aims to learn a prior policy for downstream tasks without relying on task-specific reward functions. We focus on state entropy maximization (SEM), where the goal is to learn a policy that maximizes the entropy of the state stationary distribution. In this paper, we introduce SEMDICE, a principled off-policy algorithm that computes an SEM policy from an arbitrary off-policy dataset, which optimizes the policy directly within the space of stationary distributions. SEMDICE computes a single, stationary Markov state-entropy-maximizing policy from an arbitrary off-policy dataset. Experimental results demonstrate that SEMDICE outperforms baseline algorithms in maximizing state entropy while achieving the best adaptation efficiency for downstream tasks among SEM-based unsupervised RL pre-training methods.
【7】Robust Gradient Descent via Heavy-Ball Momentum with Predictive Extrapolation
标题：通过具有预测外推的大球动量实现稳健梯度下降
链接：https://arxiv.org/abs/2512.10033
摘要：加速梯度方法，如Nesterov的加速梯度（NAG），在条件良好的问题上实现了更快的收敛，但由于积极的动量积累，在病态或非凸景观上经常发散。我们提出了重球合成梯度外推（HB-SGE），一个强大的一阶方法，结合重球动量预测梯度外推。与积累历史梯度的经典动量方法不同，HB-SGE使用局部泰勒近似来估计未来的梯度方向，在保持稳定性的同时提供自适应加速。我们证明了强凸函数的收敛保证和经验表明，HB-SGE防止发散NAG和标准动量失败的问题。在病态二次型上（条件数$κ=50$），HB-SGE在119次迭代中收敛，而SGD和NAG都发散。在非凸Rosenbrock函数上，HB-SGE在2，718次迭代中实现收敛，而经典动量方法在10步内发散。虽然NAG在条件良好的问题上仍然更快，但HB-SGE提供了一个强大的替代方案，在不同的环境中加速超过SGD，只需要O（d）$内存开销和与标准动量相同的超参数。
摘要：Accelerated gradient methods like Nesterov's Accelerated Gradient (NAG) achieve faster convergence on well-conditioned problems but often diverge on ill-conditioned or non-convex landscapes due to aggressive momentum accumulation. We propose Heavy-Ball Synthetic Gradient Extrapolation (HB-SGE), a robust first-order method that combines heavy-ball momentum with predictive gradient extrapolation. Unlike classical momentum methods that accumulate historical gradients, HB-SGE estimates future gradient directions using local Taylor approximations, providing adaptive acceleration while maintaining stability. We prove convergence guarantees for strongly convex functions and demonstrate empirically that HB-SGE prevents divergence on problems where NAG and standard momentum fail. On ill-conditioned quadratics (condition number $κ=50$), HB-SGE converges in 119 iterations while both SGD and NAG diverge. On the non-convex Rosenbrock function, HB-SGE achieves convergence in 2,718 iterations where classical momentum methods diverge within 10 steps. While NAG remains faster on well-conditioned problems, HB-SGE provides a robust alternative with speedup over SGD across diverse landscapes, requiring only $O(d)$ memory overhead and the same hyperparameters as standard momentum.
【8】UniExtreme: A Universal Foundation Model for Extreme Weather Forecasting
标题：UniExtreme：极端天气预报的通用基础模型
链接：https://arxiv.org/abs/2508.01426
备注：35 pages, 80 figures, submitted to ACM KDD 2026 conference
摘要：深度学习的最新进展导致了天气预报基础模型（FM）的发展，但它们预测极端天气事件的能力仍然有限。现有的方法要么侧重于一般的天气条件，要么专门针对特定类型的极端情况，忽视了各种极端事件的真实大气模式。在这项工作中，我们确定了极端事件的两个关键特征：（1）与正常天气状况的光谱差异，以及（2）不同极端的层次驱动因素和地理混合。沿着这条线，我们提出了UniExtreme，一个通用的极端天气预报基础模型，它集成了（1）一个自适应频率调制（AFM）模块，通过可学习的Beta分布滤波器和多粒度光谱聚合来捕获正常和极端天气之间的区域光谱差异，以及（2）事件先验增强（EPA）模块，其结合特定于区域的极端事件先验以解析分层极端多样性和复合极端模式，通过双层记忆融合网络广泛的实验表明，UniExtreme在极端和一般天气预报方面都优于最先进的基线，在各种极端情况下表现出卓越的适应性。
摘要：Recent advancements in deep learning have led to the development of Foundation Models (FMs) for weather forecasting, yet their ability to predict extreme weather events remains limited. Existing approaches either focus on general weather conditions or specialize in specific-type extremes, neglecting the real-world atmospheric patterns of diversified extreme events. In this work, we identify two key characteristics of extreme events: (1) the spectral disparity against normal weather regimes, and (2) the hierarchical drivers and geographic blending of diverse extremes. Along this line, we propose UniExtreme, a universal extreme weather forecasting foundation model that integrates (1) an Adaptive Frequency Modulation (AFM) module that captures region-wise spectral differences between normal and extreme weather, through learnable Beta-distribution filters and multi-granularity spectral aggregation, and (2) an Event Prior Augmentation (EPA) module which incorporates region-specific extreme event priors to resolve hierarchical extreme diversity and composite extreme schema, via a dual-level memory fusion network. Extensive experiments demonstrate that UniExtreme outperforms state-of-the-art baselines in both extreme and general weather forecasting, showcasing superior adaptability across diverse extreme scenarios.
【9】Deep sets and event-level maximum-likelihood estimation for fast pile-up jet rejection in ATLAS
标题：ATLAS中快速堆积喷射剔除的深集和事件级最大似然估计
链接：https://arxiv.org/abs/2512.10819
备注：6 pages, 3 figures, European Physical Society Conference on High Energy Physics (EPS-HEP2025), On behalf of the ATLAS Collaboration
摘要：在LHC的每个束团交叉处都会发生多个质子-质子碰撞（堆积），预计在运行3期间平均相互作用次数将达到80次，在高亮度LHC中达到200次。作为一个直接的结果，多射流特征的事件将以越来越高的速率发生。为了应对增加的亮度，能够有效地分组射流根据其起源沿光束线是至关重要的，特别是在触发水平。在这项工作中，一种新的不确定性感知的射流回归模型的基础上，深集架构，DIPz，回归沿光束线的射流原点位置。DIPz算法的输入是与每个射流相关联的带电粒子轨迹。一个事件级判别，最大对数乘积的似然（MLPL），构造结合DIPz每喷预测。MLPL经过切割优化，可选择与目标多喷口特征选择兼容的事件。这种组合方法提供了一种鲁棒且计算效率高的方法，用于多射流最终状态下的堆积拒绝，适用于ATLAS高水平触发器处的实时事件选择。
摘要：Multiple proton-proton collisions (pile-up) occur at every bunch crossing at the LHC, with the mean number of interactions expected to reach 80 during Run 3 and up to 200 at the High-Luminosity LHC. As a direct consequence, events with multijet signatures will occur at increasingly high rates. To cope with the increased luminosity, being able to efficiently group jets according to their origin along the beamline is crucial, particularly at the trigger level. In this work, a novel uncertainty-aware jet regression model based on a Deep Sets architecture is introduced, DIPz, to regress on a jet origin position along the beamline. The inputs to the DIPz algorithm are the charged particle tracks associated to each jet. An event-level discriminant, the Maximum Log Product of Likelihoods (MLPL), is constructed by combining the DIPz per-jet predictions. MLPL is cut-optimized to select events compatible with targeted multi-jet signature selection. This combined approach provides a robust and computationally efficient method for pile-up rejection in multi-jet final states, applicable to real-time event selections at the ATLAS High Level Trigger.
其他神经网络|深度学习|模型|建模(24篇)
【1】ImplicitRDP: An End-to-End Visual-Force Diffusion Policy with Structural Slow-Fast Learning
标题：隐含SDP：具有结构性慢-快学习的端到端视觉力扩散政策
链接：https://arxiv.org/abs/2512.10946
备注：Project page: https://implicit-rdp.github.io
摘要：人类级别的接触丰富的操纵依赖于两个关键模态的不同作用：视觉提供空间丰富但时间缓慢的全局上下文，而力感测捕获快速，高频的局部接触动态。整合这些信号是具有挑战性的，因为它们的基本频率和信息差异。在这项工作中，我们提出了ImplicitRDP，一个统一的端到端的视觉力扩散策略，集成了视觉规划和反应力控制在一个单一的网络。我们引入了结构慢-快学习，这是一种利用因果注意力同时处理异步视觉和力令牌的机制，允许策略在力频率上进行闭环调整，同时保持动作块的时间一致性。此外，为了减轻端到端模型无法在不同模态之间调整权重的模态崩溃，我们提出了基于虚拟目标的表示正则化。这个辅助目标将力反馈映射到与动作相同的空间，提供比原始力预测更强的物理基础学习信号。在接触丰富的任务上进行的大量实验表明，ImplicitRDP的性能明显优于仅视觉和分层基线，通过简化的训练管道实现了卓越的反应性和成功率。代码和视频将在https://implicit-rdp.github.io上公开。
摘要：Human-level contact-rich manipulation relies on the distinct roles of two key modalities: vision provides spatially rich but temporally slow global context, while force sensing captures rapid, high-frequency local contact dynamics. Integrating these signals is challenging due to their fundamental frequency and informational disparities. In this work, we propose ImplicitRDP, a unified end-to-end visual-force diffusion policy that integrates visual planning and reactive force control within a single network. We introduce Structural Slow-Fast Learning, a mechanism utilizing causal attention to simultaneously process asynchronous visual and force tokens, allowing the policy to perform closed-loop adjustments at the force frequency while maintaining the temporal coherence of action chunks. Furthermore, to mitigate modality collapse where end-to-end models fail to adjust the weights across different modalities, we propose Virtual-target-based Representation Regularization. This auxiliary objective maps force feedback into the same space as the action, providing a stronger, physics-grounded learning signal than raw force prediction. Extensive experiments on contact-rich tasks demonstrate that ImplicitRDP significantly outperforms both vision-only and hierarchical baselines, achieving superior reactivity and success rates with a streamlined training pipeline. Code and videos will be publicly available at https://implicit-rdp.github.io.
【2】Physics-Informed Learning of Flow Distribution and Receiver Heat Losses in Parabolic Trough Solar Fields
标题：基于物理知识的学习关于太阳能场流量分布和接收器热损失
链接：https://arxiv.org/abs/2512.10886
摘要：抛物面槽式聚光太阳能发电（CSP）厂运行大型集热器回路液压网络，尽管空间上存在光学性能、热损失和压降的异质性，但这些集热器回路必须提供均匀的出口温度。在测量回路温度的同时，无法观察回路级质量流量和接收器热损失参数，因此无法使用标准监测工具诊断液压失衡或接收器退化。 我们提出了一个物理知识的学习框架，推断（i）回路级质量流量比和（ii）直接从常规操作数据随时间变化的接收器传热系数。该方法利用夜间均质化时期-当热油通过非辐射场循环时-以隔离液压和热损失效应。可微分共轭传热模型被离散化并嵌入到端到端学习管道中，该学习管道使用来自50 MW Andasol 3太阳能场的历史工厂数据进行优化。 该模型准确地重建回路温度（RMSE $<2^\circ$C），并产生回路不平衡和接收器热损失的物理上有意义的估计。与无人机红外热成像（QScan）的比较显示出很强的对应性，正确识别了所有具有高损耗接收器的区域。这表明，嘈杂的真实世界的CSP操作数据包含足够的信息，以恢复潜在的物理参数时，结合适当的建模和可微优化。
摘要：Parabolic trough Concentrating Solar Power (CSP) plants operate large hydraulic networks of collector loops that must deliver a uniform outlet temperature despite spatially heterogeneous optical performance, heat losses, and pressure drops. While loop temperatures are measured, loop-level mass flows and receiver heat-loss parameters are unobserved, making it impossible to diagnose hydraulic imbalances or receiver degradation using standard monitoring tools. We present a physics-informed learning framework that infers (i) loop-level mass-flow ratios and (ii) time-varying receiver heat-transfer coefficients directly from routine operational data. The method exploits nocturnal homogenization periods -- when hot oil is circulated through a non-irradiated field -- to isolate hydraulic and thermal-loss effects. A differentiable conjugate heat-transfer model is discretized and embedded into an end-to-end learning pipeline optimized using historical plant data from the 50 MW Andasol 3 solar field. The model accurately reconstructs loop temperatures (RMSE $<2^\circ$C) and produces physically meaningful estimates of loop imbalances and receiver heat losses. Comparison against drone-based infrared thermography (QScan) shows strong correspondence, correctly identifying all areas with high-loss receivers. This demonstrates that noisy real-world CSP operational data contain enough information to recover latent physical parameters when combined with appropriate modeling and differentiable optimization.
【3】Generative Modeling from Black-box Corruptions via Self-Consistent Stochastic Interpolants
标题：基于自洽随机插值的黑盒腐蚀生成建模
链接：https://arxiv.org/abs/2512.10857
摘要：基于传输的方法已经成为从大型干净数据集构建生成模型的领先范例。然而，在许多科学和工程领域，干净的数据往往是不可用的：相反，我们只观察到通过噪声，病态信道损坏的测量。因此，原始数据的生成模型需要在分布水平上求解逆问题。在这项工作中，我们引入了一种新的方法来完成这项任务的基础上随机插值：我们迭代更新损坏和干净的数据样本之间的传输映射，只使用访问损坏的数据集以及黑盒访问腐败通道。在适当的条件下，这个迭代过程收敛到一个自洽的传输映射，有效地反转腐败通道，从而使生成模型的干净的数据。我们称之为自洽随机插值（SCSI）的方法。它（i）与变分替代方案相比计算效率高，（ii）高度灵活，仅用黑盒访问处理任意非线性前向模型，（iii）享有理论保证。我们在自然图像处理和科学重建中的逆问题上表现出优越的性能，并在适当的假设下建立了该方案的收敛保证。
摘要：Transport-based methods have emerged as a leading paradigm for building generative models from large, clean datasets. However, in many scientific and engineering domains, clean data are often unavailable: instead, we only observe measurements corrupted through a noisy, ill-conditioned channel. A generative model for the original data thus requires solving an inverse problem at the level of distributions. In this work, we introduce a novel approach to this task based on Stochastic Interpolants: we iteratively update a transport map between corrupted and clean data samples using only access to the corrupted dataset as well as black box access to the corruption channel. Under appropriate conditions, this iterative procedure converges towards a self-consistent transport map that effectively inverts the corruption channel, thus enabling a generative model for the clean data. We refer to the resulting method as the self-consistent stochastic interpolant (SCSI). It (i) is computationally efficient compared to variational alternatives, (ii) highly flexible, handling arbitrary nonlinear forward models with only black-box access, and (iii) enjoys theoretical guarantees. We demonstrate superior performance on inverse problems in natural image processing and scientific reconstruction, and establish convergence guarantees of the scheme under appropriate assumptions.
【4】Learning Controllable and Diverse Player Behaviors in Multi-Agent Environments
标题：多智能体环境中学习可控且多样化的玩家行为
链接：https://arxiv.org/abs/2512.10835
备注：Submitted to IEEE Transactions on Games
摘要 ：本文介绍了一种强化学习框架，它可以在不依赖人类游戏数据的情况下实现可控和多样化的玩家行为。现有的方法通常需要大规模的玩家轨迹，针对不同的玩家类型训练单独的模型，或者在可解释的行为参数和学习的策略之间不提供直接映射，从而限制了它们的可扩展性和可控性。我们在N维连续空间中定义玩家行为，并从包含代表真实人类风格的子集的区域中均匀采样目标行为向量。在训练过程中，每个智能体接收其当前和目标行为向量作为输入，奖励基于它们之间距离的归一化减少。这允许策略学习动作如何影响行为统计，从而能够平滑地控制攻击性、移动性和合作性等属性。一个单一的PPO为基础的多代理政策可以复制新的或看不见的播放风格，而无需重新训练。在自定义多人Unity游戏中进行的实验表明，所提出的框架比只赢的基线产生更大的行为多样性，并可靠地匹配不同目标的指定行为向量。该方法提供了一种可扩展的解决方案，用于自动化游戏测试，游戏平衡，类人行为模拟，以及替换在线游戏中断开连接的玩家。
摘要：This paper introduces a reinforcement learning framework that enables controllable and diverse player behaviors without relying on human gameplay data. Existing approaches often require large-scale player trajectories, train separate models for different player types, or provide no direct mapping between interpretable behavioral parameters and the learned policy, limiting their scalability and controllability. We define player behavior in an N-dimensional continuous space and uniformly sample target behavior vectors from a region that encompasses the subset representing real human styles. During training, each agent receives both its current and target behavior vectors as input, and the reward is based on the normalized reduction in distance between them. This allows the policy to learn how actions influence behavioral statistics, enabling smooth control over attributes such as aggressiveness, mobility, and cooperativeness. A single PPO-based multi-agent policy can reproduce new or unseen play styles without retraining. Experiments conducted in a custom multi-player Unity game show that the proposed framework produces significantly greater behavioral diversity than a win-only baseline and reliably matches specified behavior vectors across diverse targets. The method offers a scalable solution for automated playtesting, game balancing, human-like behavior simulation, and replacing disconnected players in online games.
【5】Beyond the Black Box: Identifiable Interpretation and Control in Generative Models via Causal Minimality
标题：超越黑匣子：通过因果极小性在生成模型中的可识别解释和控制
链接：https://arxiv.org/abs/2512.10720
摘要：深度生成模型虽然彻底改变了图像和文本生成等领域，但在很大程度上是不透明的黑匣子，阻碍了人类的理解、控制和对齐。虽然像稀疏自动编码器（SAE）这样的方法显示出显着的经验成功，但它们往往缺乏理论保证，存在主观见解的风险。我们的主要目标是为可解释的生成模型建立一个原则性的基础。我们证明了因果最小性原则-有利于最简单的因果解释-可以赋予扩散视觉和自回归语言模型的潜在表征明确的因果解释和强大的，组件明智的可识别的控制。我们介绍了一种新的理论框架分层选择模型，其中较高级别的概念出现从约束组成的较低级别的变量，更好地捕捉复杂的依赖关系，在数据生成。在理论推导的最小条件下（表现为稀疏性或压缩约束），我们表明，学习表示可以等同于数据生成过程的真实潜变量。从经验上讲，将这些约束应用于领先的生成模型，使我们能够提取其固有的层次概念图，为其内部知识组织提供新的见解。此外，这些基于因果关系的概念可以作为细粒度模型控制的杠杆，为透明、可靠的系统铺平道路。
摘要：Deep generative models, while revolutionizing fields like image and text generation, largely operate as opaque black boxes, hindering human understanding, control, and alignment. While methods like sparse autoencoders (SAEs) show remarkable empirical success, they often lack theoretical guarantees, risking subjective insights. Our primary objective is to establish a principled foundation for interpretable generative models. We demonstrate that the principle of causal minimality -- favoring the simplest causal explanation -- can endow the latent representations of diffusion vision and autoregressive language models with clear causal interpretation and robust, component-wise identifiable control. We introduce a novel theoretical framework for hierarchical selection models, where higher-level concepts emerge from the constrained composition of lower-level variables, better capturing the complex dependencies in data generation. Under theoretically derived minimality conditions (manifesting as sparsity or compression constraints), we show that learned representations can be equivalent to the true latent variables of the data-generating process. Empirically, applying these constraints to leading generative models allows us to extract their innate hierarchical concept graphs, offering fresh insights into their internal knowledge organization. Furthermore, these causally grounded concepts serve as levers for fine-grained model steering, paving the way for transparent, reliable systems.
【6】AEBNAS: Strengthening Exit Branches in Early-Exit Networks through Hardware-Aware Neural Architecture Search
标题：AEBNAS：通过硬件感知神经架构搜索加强早期退出网络中的退出分支
链接：https://arxiv.org/abs/2512.10671
摘要：早期退出网络是通过根据输入数据的复杂性调整计算来降低深度学习模型的整体能耗和延迟的有效解决方案。通过将中间出口分支并入架构中，它们为更简单的样本提供更少的计算，这对于能源消耗至关重要的资源受限设备特别有益。然而，由于需要平衡效率和性能，设计早期退出网络是一个具有挑战性和耗时的过程。最近的工作利用神经架构搜索（NAS）来设计更有效的早期退出网络，旨在通过确定架构中退出分支的最佳位置和数量来减少平均延迟，同时提高模型准确性。影响早期退出网络效率和准确性的另一个重要因素是退出分支中的层的深度和类型。在本文中，我们使用硬件感知的NAS，以加强出口分支，同时考虑在优化的准确性和效率。我们对CIFAR-10、CIFAR-100和SVHN数据集的性能评估表明，我们提出的框架考虑了出口分支的不同深度和层次以及自适应阈值调整，设计了早期出口网络，与最先进的方法相比，该网络具有相同或更低的平均MAC数，实现了更高的准确性。
摘要：Early-exit networks are effective solutions for reducing the overall energy consumption and latency of deep learning models by adjusting computation based on the complexity of input data. By incorporating intermediate exit branches into the architecture, they provide less computation for simpler samples, which is particularly beneficial for resource-constrained devices where energy consumption is crucial. However, designing early-exit networks is a challenging and time-consuming process due to the need to balance efficiency and performance. Recent works have utilized Neural Architecture Search (NAS) to design more efficient early-exit networks, aiming to reduce average latency while improving model accuracy by determining the best positions and number of exit branches in the architecture. Another important factor affecting the efficiency and accuracy of early-exit networks is the depth and types of layers in the exit branches. In this paper, we use hardware-aware NAS to strengthen exit branches, considering both accuracy and efficiency during optimization. Our performance evaluation on the CIFAR-10, CIFAR-100, and SVHN datasets demonstrates that our proposed framework, which considers varying depths and layers for exit branches along with adaptive threshold tuning, designs early-exit networks that achieve higher accuracy with the same or lower average number of MACs compared to the state-of-the-art approaches.
【7】Learning by Analogy: A Causal Framework for Composition Generalization
标题：类比学习：作文概括的因果框架
链接：https://arxiv.org/abs/2512.10669
摘要：组合泛化--理解和生成学习概念的新组合的能力--使模型能够扩展其能力，超越有限的经验。虽然有效，但实现这一关键能力的数据结构和原则仍然知之甚少。我们提出，组合泛化从根本上需要将高级概念分解为基本的低级概念，这些概念可以在类似的上下文中重新组合，类似于人类如何在概念之间进行类比。例如，从未见过孔雀吃米饭的人可以通过将其与之前观察到的鸡吃米饭相关联来想象这一场景。 在这项工作中，我们正式这些直观的过程中使用因果模块化和最小的变化的原则。我们引入了一个分层的数据生成过程，自然编码不同层次的概念和它们的相互作用机制。从理论上讲，我们证明了这种方法使组合泛化支持组成概念之间的复杂关系，超越了以前的工作，假设更简单的相互作用，如添加剂的影响。重要的是，我们还证明了这种潜在的层次结构是可证明可恢复（可识别）从可观察的数据，如文本图像对，学习这样的生成过程的必要步骤。为了验证我们的理论，我们应用我们的理论框架的见解，并在基准数据集上实现了显着的改进。
摘要 ：Compositional generalization -- the ability to understand and generate novel combinations of learned concepts -- enables models to extend their capabilities beyond limited experiences. While effective, the data structures and principles that enable this crucial capability remain poorly understood. We propose that compositional generalization fundamentally requires decomposing high-level concepts into basic, low-level concepts that can be recombined across similar contexts, similar to how humans draw analogies between concepts. For example, someone who has never seen a peacock eating rice can envision this scene by relating it to their previous observations of a chicken eating rice. In this work, we formalize these intuitive processes using principles of causal modularity and minimal changes. We introduce a hierarchical data-generating process that naturally encodes different levels of concepts and their interaction mechanisms. Theoretically, we demonstrate that this approach enables compositional generalization supporting complex relations between composed concepts, advancing beyond prior work that assumes simpler interactions like additive effects. Critically, we also prove that this latent hierarchical structure is provably recoverable (identifiable) from observable data like text-image pairs, a necessary step for learning such a generative process. To validate our theory, we apply insights from our theoretical framework and achieve significant improvements on benchmark datasets.
【8】Authority Backdoor: A Certifiable Backdoor Mechanism for Authoring DNNs
标题：权威后门：一种用于编写DNN的可认证后门机制
链接：https://arxiv.org/abs/2512.10600
备注：Accepted to AAAI 2026 (Main Track). Code is available at: https://github.com/PlayerYangh/Authority-Trigger
摘要：深度神经网络（DNN）作为有价值的知识产权，面临未经授权的使用。现有的保护措施，如数字水印，在很大程度上是被动的;它们只提供事后所有权验证，不能主动防止非法使用被盗模型。这项工作提出了一个积极的保护计划，被称为“权威后门”，它直接嵌入到模型的访问限制。特别地，该方案利用后门学习框架来内在地锁定模型的效用，使得它仅在存在特定触发器（例如，硬件指纹）。但如果没有它，DNN的性能就会下降到无用。为了进一步提高安全性，建议的权威机构计划，可证明的鲁棒性集成，以防止自适应攻击者删除植入的后门。由此产生的框架为DNN建立了一个安全的权威机制，将访问控制与对抗性攻击的可认证鲁棒性相结合。在不同的架构和数据集上进行的大量实验验证了所提出的框架的有效性和可证明的鲁棒性。
摘要：Deep Neural Networks (DNNs), as valuable intellectual property, face unauthorized use. Existing protections, such as digital watermarking, are largely passive; they provide only post-hoc ownership verification and cannot actively prevent the illicit use of a stolen model. This work proposes a proactive protection scheme, dubbed ``Authority Backdoor," which embeds access constraints directly into the model. In particular, the scheme utilizes a backdoor learning framework to intrinsically lock a model's utility, such that it performs normally only in the presence of a specific trigger (e.g., a hardware fingerprint). But in its absence, the DNN's performance degrades to be useless. To further enhance the security of the proposed authority scheme, the certifiable robustness is integrated to prevent an adaptive attacker from removing the implanted backdoor. The resulting framework establishes a secure authority mechanism for DNNs, combining access control with certifiable robustness against adversarial attacks. Extensive experiments on diverse architectures and datasets validate the effectiveness and certifiable robustness of the proposed framework.
【9】Is the Information Bottleneck Robust Enough? Towards Label-Noise Resistant Information Bottleneck Learning
标题：信息瓶颈足够强大吗？迈向抗标签噪音的信息瓶颈学习
链接：https://arxiv.org/abs/2512.10573
备注：Accepted by the Main Technical Track of the 40th Annual AAAI Conference on Artificial Intelligence (AAAI-2026)
摘要：信息瓶颈（IB）原则通过保留标签相关信息同时压缩不相关信息来促进有效的表征学习。然而，它对准确标签的强烈依赖使其本质上容易受到标签噪声的影响，这在现实世界中很普遍，导致显着的性能下降和过拟合。为了解决这个问题，我们提出了LaT-IB，一种新的标签抗噪信息瓶颈方法，它引入了“最小足够清洁”（MSC）标准。作为一个互信息正则化保留任务相关的信息，同时丢弃噪音，MSC解决标准IB的脆弱性，嘈杂的标签监督。为了实现这一点，LaT-IB采用了噪声感知的潜在解纠缠，将潜在表示分解为与干净标签空间和噪声空间对齐的分量。从理论上讲，我们首先推导出我们的目标，包括预测，压缩和解纠缠的每个组件的互信息界，而且证明了优化它鼓励表示不变的输入噪声和分离干净和嘈杂的标签信息。此外，我们设计了一个三阶段的训练框架：热身，知识注入和鲁棒训练，逐步引导模型向抗噪表示。大量实验表明，LaT-IB在标签噪声下实现了卓越的鲁棒性和效率，显着增强了标签噪声的真实世界场景中的鲁棒性和适用性。
摘要：The Information Bottleneck (IB) principle facilitates effective representation learning by preserving label-relevant information while compressing irrelevant information. However, its strong reliance on accurate labels makes it inherently vulnerable to label noise, prevalent in real-world scenarios, resulting in significant performance degradation and overfitting. To address this issue, we propose LaT-IB, a novel Label-Noise ResistanT Information Bottleneck method which introduces a "Minimal-Sufficient-Clean" (MSC) criterion. Instantiated as a mutual information regularizer to retain task-relevant information while discarding noise, MSC addresses standard IB's vulnerability to noisy label supervision. To achieve this, LaT-IB employs a noise-aware latent disentanglement that decomposes the latent representation into components aligned with to the clean label space and the noise space. Theoretically, we first derive mutual information bounds for each component of our objective including prediction, compression, and disentanglement, and moreover prove that optimizing it encourages representations invariant to input noise and separates clean and noisy label information. Furthermore, we design a three-phase training framework: Warmup, Knowledge Injection and Robust Training, to progressively guide the model toward noise-resistant representations. Extensive experiments demonstrate that LaT-IB achieves superior robustness and efficiency under label noise, significantly enhancing robustness and applicability in real-world scenarios with label noise.
【10】Mode-Seeking for Inverse Problems with Diffusion Models
标题：扩散模型反问题的寻模
链接：https://arxiv.org/abs/2512.10524
摘要：预训练的无条件扩散模型，结合后验采样或最大后验（MAP）估计技术，可以解决任意逆问题，而无需特定任务的训练或微调。然而，现有的后验采样和MAP估计方法通常依赖于建模近似，并且计算量很大。在这项工作中，我们提出了变分寻模损失（VML），当在每个反向扩散步骤中最小化时，引导生成的样本朝向MAP估计。VML产生于最小化扩散后验$p（\mathbf{x}_0）之间的Kullback-Leibler（KL）发散的新视角|\mathbf{x}_t）$和测量后验$p（\mathbf{x}_0|\mathbf{y}）$，其中$\mathbf{y}$表示测量值。重要的是，对于线性逆问题，VML可以解析导出，不需要近似。基于进一步的理论见解，我们提出了VML-MAP，一个经验有效的算法来解决反问题，并验证其有效性在现有的方法在性能和计算时间，通过广泛的实验在多个数据集上的各种图像恢复任务。
摘要：A pre-trained unconditional diffusion model, combined with posterior sampling or maximum a posteriori (MAP) estimation techniques, can solve arbitrary inverse problems without task-specific training or fine-tuning. However, existing posterior sampling and MAP estimation methods often rely on modeling approximations and can be computationally demanding. In this work, we propose the variational mode-seeking loss (VML), which, when minimized during each reverse diffusion step, guides the generated sample towards the MAP estimate. VML arises from a novel perspective of minimizing the Kullback-Leibler (KL) divergence between the diffusion posterior $p(\mathbf{x}_0|\mathbf{x}_t)$ and the measurement posterior $p(\mathbf{x}_0|\mathbf{y})$, where $\mathbf{y}$ denotes the measurement. Importantly, for linear inverse problems, VML can be analytically derived and need not be approximated. Based on further theoretical insights, we propose VML-MAP, an empirically effective algorithm for solving inverse problems, and validate its efficacy over existing methods in both performance and computational time, through extensive experiments on diverse image-restoration tasks across multiple datasets.
【11】T-SKM-Net: Trainable Neural Network Framework for Linear Constraint Satisfaction via Sampling Kaczmarz-Motzkin Method
标题
：T-SKM-Net：通过采样Kaczmarz-Motzkin方法实现线性约束满足的可训练神经网络框架
链接：https://arxiv.org/abs/2512.10461
摘要：神经网络约束满足对于电力系统优化、机器人路径规划和自动驾驶等安全关键型应用至关重要。然而，现有的约束满足方法面临效率和适用性的权衡，硬约束方法要么计算复杂度高，要么约束结构的限制性假设。Sampling Kaczmarz-Motzkin（SKM）方法是一种用于求解大型线性不等式系统的随机迭代算法，具有良好的收敛性，但其argmax运算引入了不可微性，给神经网络应用带来了挑战。这项工作提出了可训练采样Kaczmarz-Motzkin网络（T-SKM-Net）框架，并首次将SKM类型的方法系统地集成到神经网络约束满足中。该框架通过零空间变换将混合约束问题转化为纯不等式问题，采用SKM迭代求解，并将解映射回原始约束空间，有效地处理等式和不等式约束。我们提供了基于无偏梯度估计的期望和端到端可训练性保证的后处理有效性的理论证明，表明尽管不可微操作，该框架支持标准的反向传播。在DCOPF case118基准测试中，我们的方法在后处理模式下实现了4.27ms/项GPU串行前向推理，最大最优间隙为0.0025%，在联合训练模式下实现了5.25ms/项GPU串行前向推理，最大最优间隙为0.0008%，与pandapower求解器相比，提供了超过25$\times $的加速比，同时在给定容差下保持零约束违反。
摘要：Neural network constraint satisfaction is crucial for safety-critical applications such as power system optimization, robotic path planning, and autonomous driving. However, existing constraint satisfaction methods face efficiency-applicability trade-offs, with hard constraint methods suffering from either high computational complexity or restrictive assumptions on constraint structures. The Sampling Kaczmarz-Motzkin (SKM) method is a randomized iterative algorithm for solving large-scale linear inequality systems with favorable convergence properties, but its argmax operations introduce non-differentiability, posing challenges for neural network applications. This work proposes the Trainable Sampling Kaczmarz-Motzkin Network (T-SKM-Net) framework and, for the first time, systematically integrates SKM-type methods into neural network constraint satisfaction. The framework transforms mixed constraint problems into pure inequality problems through null space transformation, employs SKM for iterative solving, and maps solutions back to the original constraint space, efficiently handling both equality and inequality constraints. We provide theoretical proof of post-processing effectiveness in expectation and end-to-end trainability guarantees based on unbiased gradient estimators, demonstrating that despite non-differentiable operations, the framework supports standard backpropagation. On the DCOPF case118 benchmark, our method achieves 4.27ms/item GPU serial forward inference with 0.0025% max optimality gap with post-processing mode and 5.25ms/item with 0.0008% max optimality gap with joint training mode, delivering over 25$\times$ speedup compared to the pandapower solver while maintaining zero constraint violations under given tolerance.
【12】Metacognitive Sensitivity for Test-Time Dynamic Model Selection
标题：测试时动态模型选择的元认知敏感性
链接：https://arxiv.org/abs/2512.10451
备注：Accepted at the NeurIPS 2025 CogInterp Workshop
摘要：人类认知的一个关键方面是元认知-评估自己的知识和判断可靠性的能力。虽然深度学习模型可以表达对预测的信心，但它们通常会受到校准不良的影响，这是一种认知偏差，其中表达的信心并不反映真正的能力。模特们真的知道他们知道什么吗？借鉴人类认知科学，我们提出了一个评估和利用AI元认知的新框架。我们引入元d '，一种基于心理学的元认知敏感性测量，来评估模型的置信度预测其自身准确性的可靠性。然后，我们使用这个动态的灵敏度得分作为上下文的强盗为基础的仲裁器，执行测试时的模型选择，学习几个专家模型信任一个给定的任务。我们在多个数据集和深度学习模型组合（包括CNN和VLM）上的实验表明，这种元认知方法提高了组成模型的联合推理准确性。这项工作提供了一种新的人工智能模型的行为解释，将集成选择重新定义为评估短期信号（置信度预测分数）和中期特征（元认知敏感性）的问题。
摘要：A key aspect of human cognition is metacognition - the ability to assess one's own knowledge and judgment reliability. While deep learning models can express confidence in their predictions, they often suffer from poor calibration, a cognitive bias where expressed confidence does not reflect true competence. Do models truly know what they know? Drawing from human cognitive science, we propose a new framework for evaluating and leveraging AI metacognition. We introduce meta-d', a psychologically-grounded measure of metacognitive sensitivity, to characterise how reliably a model's confidence predicts its own accuracy. We then use this dynamic sensitivity score as context for a bandit-based arbiter that performs test-time model selection, learning which of several expert models to trust for a given task. Our experiments across multiple datasets and deep learning model combinations (including CNNs and VLMs) demonstrate that this metacognitive approach improves joint-inference accuracy over constituent models. This work provides a novel behavioural account of AI models, recasting ensemble selection as a problem of evaluating both short-term signals (confidence prediction scores) and medium-term traits (metacognitive sensitivity).
【13】The Operator Origins of Neural Scaling Laws: A Generalized Spectral Transport Dynamics of Deep Learning
标题：神经缩放定律的运算符起源：深度学习的广义谱传输动力学
链接：https://arxiv.org/abs/2512.10427
摘要：现代深度网络在粗糙的有限规则性区域中运行，其中雅可比诱导算子表现出重尾谱和强基漂移。在这项工作中，我们得到了一个统一的算子theoretoretic描述的神经训练动力学直接从梯度下降。从函数空间中的精确演化$\dot e_t = -M（t）e_t$出发，应用Kato微扰理论得到了一个严格的耦合模常微分方程组，并证明了在粗粒化后，这些动力学收敛到一个谱输运耗散偏微分方程\[ \partial_tg + \partial_λ（vg）= -λg + S，\]其中$v$捕获本征基漂移，$S$编码非局部谱耦合。 我们证明了神经训练保持了函数的规律性，迫使漂移采取渐近幂律形式$v（λ，t）\sim -c（t）λ^b$。在弱耦合制度-自然引起的频谱局部性和SGD噪声-PDE承认自相似的解决方案，分辨率前沿，多项式振幅增长，和幂律耗散。这种结构产生了显式的标度律指数，解释了双重下降的几何结构，并表明对于缓慢变化的$L$，有效训练时间满足$τ（t）=t^αL（t）$。 最后，我们证明了NTK训练和特征学习是由于同一PDE的两个限制而产生的：$v\equiv 0$恢复了懒惰的动态，而$v\neq 0$产生了表示漂移。我们的研究结果提供了一个统一的光谱框架，连接算子几何，优化动力学和现代深度网络的通用缩放行为。
摘要：Modern deep networks operate in a rough, finite-regularity regime where Jacobian-induced operators exhibit heavy-tailed spectra and strong basis drift. In this work, we derive a unified operator-theoretoretic description of neural training dynamics directly from gradient descent. Starting from the exact evolution $\dot e_t = -M(t)e_t$ in function space, we apply Kato perturbation theory to obtain a rigorous system of coupled mode ODEs and show that, after coarse-graining, these dynamics converge to a spectral transport--dissipation PDE \[ \partial_t g + \partial_λ(v g) = -λg + S, \] where $v$ captures eigenbasis drift and $S$ encodes nonlocal spectral coupling. We prove that neural training preserves functional regularity, forcing the drift to take an asymptotic power-law form $v(λ,t)\sim -c(t)λ^b$. In the weak-coupling regime -- naturally induced by spectral locality and SGD noise -- the PDE admits self-similar solutions with a resolution frontier, polynomial amplitude growth, and power-law dissipation. This structure yields explicit scaling-law exponents, explains the geometry of double descent, and shows that the effective training time satisfies $τ(t)=t^αL(t)$ for slowly varying $L$. Finally, we show that NTK training and feature learning arise as two limits of the same PDE: $v\equiv 0$ recovers lazy dynamics, while $v\neq 0$ produces representation drift. Our results provide a unified spectral framework connecting operator geometry, optimization dynamics, and the universal scaling behavior of modern deep networks.
【14】RoboNeuron: A Modular Framework Linking Foundation Models and ROS for Embodied AI
标题：RoboNeuron：连接基础模型和可编程人工智能的Ro模块化框架
链接：https://arxiv.org/abs/2512.10394
摘要：目前的嵌入式AI系统面临着严重的工程障碍，主要表现为跨场景适应性差，模块间耦合刚性，以及碎片化的推理加速。为了克服这些限制，我们提出了RoboNeuron，一个通用的部署框架体现的智能。RoboNeuron是第一个将大型语言模型（LLM）和视觉-语言-动作（VLA）模型的认知能力与机器人操作系统（ROS）的实时执行骨干深度集成的框架。我们利用模型上下文协议（MCP）作为语义桥梁，使LLM动态编排底层机器人工具。该框架建立了一个高度模块化的体系结构，通过利用ROS的统一通信接口，严格实现了传感、推理和控制。至关重要的是，我们引入了一个自动化工具来将ROS消息转换为可调用的MCP函数，从而大大简化了开发。RoboNeuron显著增强了跨场景适应性和组件灵活性，同时为水平性能基准测试建立了系统平台，为可扩展的现实世界体现应用奠定了坚实的基础。
摘要：Current embodied AI systems face severe engineering impediments, primarily characterized by poor cross-scenario adaptability, rigid inter-module coupling, and fragmented inference acceleration. To overcome these limitations, we propose RoboNeuron, a universal deployment framework for embodied intelligence. RoboNeuron is the first framework to deeply integrate the cognitive capabilities of Large Language Models (LLMs) and Vision-Language-Action (VLA) models with the real-time execution backbone of the Robot Operating System (ROS). We utilize the Model Context Protocol (MCP) as a semantic bridge, enabling the LLM to dynamically orchestrate underlying robotic tools. The framework establishes a highly modular architecture that strictly decouples sensing, reasoning, and control by leveraging ROS's unified communication interfaces. Crucially, we introduce an automated tool to translate ROS messages into callable MCP functions, significantly streamlining development. RoboNeuron significantly enhances cross-scenario adaptability and component flexibility, while establishing a systematic platform for horizontal performance benchmarking, laying a robust foundation for scalable real-world embodied applications.
【15】D2M: A Decentralized, Privacy-Preserving, Incentive-Compatible Data Marketplace for Collaborative Learning
标题：D2 M：一个去中心化、隐私保护、激励兼容的协作学习数据市场
链接：https://arxiv.org/abs/2512.10372
摘要：对协作机器学习和数据分析的需求不断增长，需要安全和分散的数据共享框架来平衡隐私，信任和激励。现有的方法，包括联邦学习（FL）和基于区块链的数据市场，都存在不足：FL通常依赖于可信的聚合器，缺乏拜占庭式的鲁棒性，而区块链框架则难以进行计算密集型的培训和激励集成。 我们介绍了\prot，一个去中心化的数据市场，它将联邦学习，区块链仲裁和经济激励统一到一个保护隐私的数据共享框架中。“prot”使数据买家能够通过区块链智能合约提交基于投标的请求，该合约管理拍卖，托管和争议解决。计算密集型训练被委托给\cone\（\uline{E} xtask的\uline{Co}mpute \uline{N}网络），这是一个链下分布式执行层。为了防止对抗行为，\prot\将修改后的YODA协议与指数增长的执行集集成在一起，以实现弹性共识，并引入了Corrected OSMD来减少来自卖家的恶意或低质量贡献。所有的协议都是激励兼容的，我们的博弈论分析建立诚实为主导策略。 我们在以太坊上实现了\prot\，并在不同的对抗设置下，在基准数据集- MNIST，Fashion-MNIST和CIFAR-10上对其进行了评估。prot在MNIST和Fashion-MNIST上分别达到了99%和90%的准确率，在30%的拜占庭节点上的性能下降不到3%，在CIFAR-10上的准确率为56%，尽管它很复杂。我们的研究结果表明，“prot”可以确保隐私，在对抗条件下保持鲁棒性，并随着参与者的数量有效地扩展，使其成为现实世界分散数据共享的实用基础。
摘要：The rising demand for collaborative machine learning and data analytics calls for secure and decentralized data sharing frameworks that balance privacy, trust, and incentives. Existing approaches, including federated learning (FL) and blockchain-based data markets, fall short: FL often depends on trusted aggregators and lacks Byzantine robustness, while blockchain frameworks struggle with computation-intensive training and incentive integration. We present \prot, a decentralized data marketplace that unifies federated learning, blockchain arbitration, and economic incentives into a single framework for privacy-preserving data sharing. \prot\ enables data buyers to submit bid-based requests via blockchain smart contracts, which manage auctions, escrow, and dispute resolution. Computationally intensive training is delegated to \cone\ (\uline{Co}mpute \uline{N}etwork for \uline{E}xecution), an off-chain distributed execution layer. To safeguard against adversarial behavior, \prot\ integrates a modified YODA protocol with exponentially growing execution sets for resilient consensus, and introduces Corrected OSMD to mitigate malicious or low-quality contributions from sellers. All protocols are incentive-compatible, and our game-theoretic analysis establishes honesty as the dominant strategy. We implement \prot\ on Ethereum and evaluate it over benchmark datasets -- MNIST, Fashion-MNIST, and CIFAR-10 -- under varying adversarial settings. \prot\ achieves up to 99\% accuracy on MNIST and 90\% on Fashion-MNIST, with less than 3\% degradation up to 30\% Byzantine nodes, and 56\% accuracy on CIFAR-10 despite its complexity. Our results show that \prot\ ensures privacy, maintains robustness under adversarial conditions, and scales efficiently with the number of participants, making it a practical foundation for real-world decentralized data sharing.
【16】A Privacy-Preserving Cloud Architecture for Distributed Machine Learning at Scale
标题：用于大规模分布式机器学习的保护隐私的云架构
链接：https://arxiv.org/abs/2512.10341
摘要：分布式机器学习系统需要强大的隐私保证、可验证的合规性以及跨异构和多云环境的可扩展部署。这项工作引入了一个云原生隐私保护架构，该架构集成了联邦学习、差分隐私、零知识合规性证明以及由强化学习提供支持的自适应治理。该框架支持安全的模型训练和推理，而无需集中敏感数据，同时支持跨机构和云平台的加密可验证策略执行。跨混合Kubernetes集群部署的完整原型证明了降低的成员推断风险，一致执行正式的隐私预算，以及在差异隐私下的稳定模型性能。跨多机构工作负载的实验评估表明，该架构以最小的开销保持实用性，同时提供持续的风险感知治理。提出的框架为大规模部署可信和合规的分布式机器学习系统奠定了实践基础。
摘要：Distributed machine learning systems require strong privacy guarantees, verifiable compliance, and scalable deploy- ment across heterogeneous and multi-cloud environments. This work introduces a cloud-native privacy-preserving architecture that integrates federated learning, differential privacy, zero- knowledge compliance proofs, and adaptive governance powered by reinforcement learning. The framework supports secure model training and inference without centralizing sensitive data, while enabling cryptographically verifiable policy enforcement across institutions and cloud platforms. A full prototype deployed across hybrid Kubernetes clusters demonstrates reduced membership- inference risk, consistent enforcement of formal privacy budgets, and stable model performance under differential privacy. Ex- perimental evaluation across multi-institution workloads shows that the architecture maintains utility with minimal overhead while providing continuous, risk-aware governance. The pro- posed framework establishes a practical foundation for deploying trustworthy and compliant distributed machine learning systems at scale.
【17】The 2025 Foundation Model Transparency Index
标题：2025年基金会模型透明度指数
链接：https://arxiv.org/abs/2512.10169
备注：Website: https://crfm.stanford.edu/fmti/December-2025/index.html
摘要：基础模型开发商是世界上最重要的公司之一。随着这些公司变得越来越重要，它们的透明度做法如何演变？2025年基金会模型透明度指数是年度工作的第三版，旨在描述和量化基金会模型开发人员的透明度。2025年FMTI引入了与数据采集、使用数据和监控相关的新指标，并首次对阿里巴巴、DeepSeek和xAI等公司进行了评估。2024年FMTI报告称透明度正在提高，但2025年FMTI发现这一进展有所恶化：100分的平均得分从2024年的58分下降到2025年的40分。公司对他们的训练数据和训练计算以及部署后的使用和旗舰模型的影响最不透明。尽管总体趋势如此，IBM仍然是一个积极的离群值，得分为95，而得分最低的xAI和Midjourney仅为14。我们对Frontier Model Forum的五个成员进行了评分，结果排名居中：我们认为，这些公司避免了因得分低而造成的声誉损害，但缺乏成为透明度领导者的激励。随着世界各地的政策制定者越来越多地要求某些类型的透明度，这项工作揭示了基金会模型开发人员的透明度现状，它可能会如何改变新颁布的政策，以及需要更积极的政策干预来解决关键的信息赤字。
摘要：Foundation model developers are among the world's most important companies. As these companies become increasingly consequential, how do their transparency practices evolve? The 2025 Foundation Model Transparency Index is the third edition of an annual effort to characterize and quantify the transparency of foundation model developers. The 2025 FMTI introduces new indicators related to data acquisition, usage data, and monitoring and evaluates companies like Alibaba, DeepSeek, and xAI for the first time. The 2024 FMTI reported that transparency was improving, but the 2025 FMTI finds this progress has deteriorated: the average score out of 100 fell from 58 in 2024 to 40 in 2025. Companies are most opaque about their training data and training compute as well as the post-deployment usage and impact of their flagship models. In spite of this general trend, IBM stands out as a positive outlier, scoring 95, in contrast to the lowest scorers, xAI and Midjourney, at just 14. The five members of the Frontier Model Forum we score end up in the middle of the Index: we posit that these companies avoid reputational harms from low scores but lack incentives to be transparency leaders. As policymakers around the world increasingly mandate certain types of transparency, this work reveals the current state of transparency for foundation model developers, how it may change given newly enacted policy, and where more aggressive policy interventions are necessary to address critical information deficits.
【18】Latent Action World Models for Control with Unlabeled Trajectories
标题：具有未标记轨迹的控制的潜在行动世界模型
链接：https://arxiv.org/abs/2512.10016
摘要：受人类如何将直接互动与无动作体验相结合的启发（例如，视频），我们研究从异构数据中学习的世界模型。标准世界模型通常依赖于行动条件轨迹，这在行动标签稀缺时限制了有效性。我们介绍了一个家庭的潜在行动的世界模型，共同使用行动的条件和行动自由的数据，通过学习一个共享的潜在行动表示。这个潜在空间将观察到的控制信号与从被动观察推断的动作对齐，使得单个动力学模型能够在大规模未标记的轨迹上进行训练，同时只需要一小组标记动作的轨迹。我们使用潜伏动作世界模型通过离线强化学习（RL）来学习潜伏动作策略，从而桥接两个传统上独立的领域：离线RL（通常依赖于动作条件数据）和无动作训练（很少用于后续RL）。在DeepMind Control Suite上，我们的方法实现了强大的性能，同时使用的动作标记样本比纯粹的动作条件基线少一个数量级。这些结果表明，潜在动作可以在被动和交互数据上进行训练，从而使世界模型更有效地学习。
摘要：Inspired by how humans combine direct interaction with action-free experience (e.g., videos), we study world models that learn from heterogeneous data. Standard world models typically rely on action-conditioned trajectories, which limits effectiveness when action labels are scarce. We introduce a family of latent-action world models that jointly use action-conditioned and action-free data by learning a shared latent action representation. This latent space aligns observed control signals with actions inferred from passive observations, enabling a single dynamics model to train on large-scale unlabeled trajectories while requiring only a small set of action-labeled ones. We use the latent-action world model to learn a latent-action policy through offline reinforcement learning (RL), thereby bridging two traditionally separate domains: offline RL, which typically relies on action-conditioned data, and action-free training, which is rarely used with subsequent RL. On the DeepMind Control Suite, our approach achieves strong performance while using about an order of magnitude fewer action-labeled samples than purely action-conditioned baselines. These results show that latent actions enable training on both passive and interactive data, which makes world models learn more efficiently.
【19】Noisy Quantum Learning Theory
标题：噪音量子学习理论
链接：https://arxiv.org/abs/2512.10929
备注：11+53 pages, 3 figures
摘要：我们开发了一个从噪声量子实验中学习的框架，重点是通过噪声耦合访问未表征系统的容错设备。我们的出发点是复杂性类$\textsf{NBQP}$（“噪声BQP”），建模噪声容错量子计算机，通常不能纠正它们查询的oracle系统的错误。使用这个类，我们表明，对于自然的预言机问题，噪声可以消除指数量子学习的优势，理想的无噪声学习者，同时保持NISQ和容错设备之间的超多项式差距。除了甲骨文分离，我们研究具体的嘈杂的学习任务。对于纯度测试，指数双拷贝优势在局部去极化噪声的单次应用下崩溃。尽管如此，我们确定了一个由AdS/CFT激励的设置，其中噪声弹性结构在嘈杂的制度中恢复了量子学习优势。然后，我们分析嘈杂的泡利阴影层析成像，推导出下限，表征实例大小，量子内存，和噪声控制样本的复杂性，并设计算法与参数相似的缩放。总之，我们的研究结果表明，贝尔基础和SWAP测试原语的基础上，大多数指数量子学习的优势是从根本上脆弱的噪声，除非实验系统具有潜在的噪声鲁棒结构。因此，在未来的实验中实现有意义的量子优势将需要了解噪声鲁棒物理特性如何与可用的算法技术相结合。
摘要：We develop a framework for learning from noisy quantum experiments, focusing on fault-tolerant devices accessing uncharacterized systems through noisy couplings. Our starting point is the complexity class $\textsf{NBQP}$ ("noisy BQP"), modeling noisy fault-tolerant quantum computers that cannot, in general, error-correct the oracle systems they query. Using this class, we show that for natural oracle problems, noise can eliminate exponential quantum learning advantages of ideal noiseless learners while preserving a superpolynomial gap between NISQ and fault-tolerant devices. Beyond oracle separations, we study concrete noisy learning tasks. For purity testing, the exponential two-copy advantage collapses under a single application of local depolarizing noise. Nevertheless, we identify a setting motivated by AdS/CFT in which noise-resilient structure restores a quantum learning advantage in a noisy regime. We then analyze noisy Pauli shadow tomography, deriving lower bounds that characterize how instance size, quantum memory, and noise control sample complexity, and design algorithms with parametrically similar scalings. Together, our results show that the Bell-basis and SWAP-test primitives underlying most exponential quantum learning advantages are fundamentally fragile to noise unless the experimental system has latent noise-robust structure. Thus, realizing meaningful quantum advantages in future experiments will require understanding how noise-robust physical properties interface with available algorithmic techniques.
【20】Flexible Deep Neural Networks for Partially Linear Survival Data
标题：用于部分线性生存数据的灵活深度神经网络
链接：https://arxiv.org/abs/2512.10570
摘要：我们提出了一个灵活的深度神经网络（DNN）框架，用于在部分线性回归结构中对生存数据进行建模。该方法通过主要关注的协变量的参数线性分量保持可解释性，而非参数DNN分量捕获滋扰变量之间的复杂时间协变量相互作用。我们将该方法称为FLEXI-Haz，这是一种具有部分线性结构的灵活风险模型。与部分线性Cox模型的现有DNN方法相比，FLEXI-Haz不依赖于比例风险假设。我们建立理论保证：神经网络组件达到最小最大最优的收敛速度的基础上复合持有人类，和线性估计是根-n一致的，渐近正常的，和半参数有效的。广泛的模拟和真实数据分析表明，FLEXI-Haz提供了线性效应的准确估计，为基于比例风险的现代方法提供了一种原则性和可解释的替代方案。用于实现FLEXI-Haz的代码以及用于再现数据分析和模拟的脚本可在以下网址获得：https://github.com/AsafBanana/FLEXI-Haz
摘要 ：We propose a flexible deep neural network (DNN) framework for modeling survival data within a partially linear regression structure. The approach preserves interpretability through a parametric linear component for covariates of primary interest, while a nonparametric DNN component captures complex time-covariate interactions among nuisance variables. We refer to the method as FLEXI-Haz, a flexible hazard model with a partially linear structure. In contrast to existing DNN approaches for partially linear Cox models, FLEXI-Haz does not rely on the proportional hazards assumption. We establish theoretical guarantees: the neural network component attains minimax-optimal convergence rates based on composite Holder classes, and the linear estimator is root-n consistent, asymptotically normal, and semiparametrically efficient. Extensive simulations and real-data analyses demonstrate that FLEXI-Haz provides accurate estimation of the linear effect, offering a principled and interpretable alternative to modern methods based on proportional hazards. Code for implementing FLEXI-Haz, as well as scripts for reproducing data analyses and simulations, is available at: https://github.com/AsafBanana/FLEXI-Haz
【21】Tracking large chemical reaction networks and rare events by neural networks
标题：利用神经网络跟踪大型化学反应网络和罕见事件
链接：https://arxiv.org/abs/2512.10309
摘要：化学反应网络广泛用于化学动力学、系统生物学和流行病学中的随机动力学建模。由于大的状态空间随系统大小呈指数增长，求解控制这些系统的化学主方程构成了重大挑战。自回归神经网络的发展为这个问题提供了一个灵活的框架;然而，它的效率是有限的，特别是对于高维系统和罕见事件的情况下。在这里，我们通过利用更快的优化（如自然梯度下降和时间相关变分原理）来推动神经网络方法的前沿，实现5到22倍的加速，并利用增强采样策略来捕获罕见事件。我们证明了降低计算成本和更高的精度比以前的神经网络方法在具有挑战性的反应网络，包括丝裂原活化蛋白激酶（MAPK）级联网络，迄今为止最大的生物网络处理的解决化学主方程的方法。我们进一步将这种方法应用于空间扩展的反应扩散系统，具有罕见事件的Schlögl模型，在二维晶格上，超越了最近处理一维晶格的张量网络方法。因此，本方法使得一般化学反应网络的有效建模成为可能。
摘要：Chemical reaction networks are widely used to model stochastic dynamics in chemical kinetics, systems biology and epidemiology. Solving the chemical master equation that governs these systems poses a significant challenge due to the large state space exponentially growing with system sizes. The development of autoregressive neural networks offers a flexible framework for this problem; however, its efficiency is limited especially for high-dimensional systems and in scenarios with rare events. Here, we push the frontier of neural-network approach by exploiting faster optimizations such as natural gradient descent and time-dependent variational principle, achieving a 5- to 22-fold speedup, and by leveraging enhanced-sampling strategies to capture rare events. We demonstrate reduced computational cost and higher accuracy over the previous neural-network method in challenging reaction networks, including the mitogen-activated protein kinase (MAPK) cascade network, the hitherto largest biological network handled by the previous approaches of solving the chemical master equation. We further apply the approach to spatially extended reaction-diffusion systems, the Schlögl model with rare events, on two-dimensional lattices, beyond the recent tensor-network approach that handles one-dimensional lattices. The present approach thus enables efficient modeling of chemical reaction networks in general.
【22】Galaxy Phase-Space and Field-Level Cosmology: The Strength of Semi-Analytic Models
标题：银河相空间和场级宇宙学：半分析模型的力量
链接：https://arxiv.org/abs/2512.10222
备注：23 pages, 5 figures
摘要：半解析模型是在宇宙学框架内模拟星系性质的一种广泛使用的方法，依赖于简化但物理动机的处方。它们也被证明是生成准确星系目录的有效替代方案，与完整的流体动力学模拟相比，提供了更快，计算成本更低的选择。在本文中，我们证明，仅使用星系$3$D的位置和径向速度，我们可以训练一个图形神经网络耦合到一个时刻神经网络，以获得一个强大的机器学习为基础的模型，能够估计的物质密度参数，$Ω_{\rm}$，精度约为10%。该网络是在L-Galerkin的星系目录（25 h^{-1}$Mpc）$^3$卷上训练的，可以成功地将其预测外推到其他半解析模型（GAEA，SC-SAM和Shark），更值得注意的是，流体动力学模拟（Astrid，SIMBA，IllustrisTNG和SWIFT-EAGLE）。我们的研究结果表明，该网络是强大的天体物理和次网格物理，宇宙学和天体物理参数的变化，以及不同的halo-profile处理跨模拟。这表明，半解析模型的相空间中编码的物理关系在很大程度上独立于其特定的物理处方，加强了它们作为宇宙学参数推断的现实模拟目录生成工具的潜力。
摘要：Semi-analytic models are a widely used approach to simulate galaxy properties within a cosmological framework, relying on simplified yet physically motivated prescriptions. They have also proven to be an efficient alternative for generating accurate galaxy catalogs, offering a faster and less computationally expensive option compared to full hydrodynamical simulations. In this paper, we demonstrate that using only galaxy $3$D positions and radial velocities, we can train a graph neural network coupled to a moment neural network to obtain a robust machine learning based model capable of estimating the matter density parameters, $Ω_{\rm m}$, with a precision of approximately 10%. The network is trained on ($25 h^{-1}$Mpc)$^3$ volumes of galaxy catalogs from L-Galaxies and can successfully extrapolate its predictions to other semi-analytic models (GAEA, SC-SAM, and Shark) and, more remarkably, to hydrodynamical simulations (Astrid, SIMBA, IllustrisTNG, and SWIFT-EAGLE). Our results show that the network is robust to variations in astrophysical and subgrid physics, cosmological and astrophysical parameters, and the different halo-profile treatments used across simulations. This suggests that the physical relationships encoded in the phase-space of semi-analytic models are largely independent of their specific physical prescriptions, reinforcing their potential as tools for the generation of realistic mock catalogs for cosmological parameter inference.
【23】On Learning-Curve Monotonicity for Maximum Likelihood Estimators
标题：最大似然估计量的学习曲线单调性
链接：https://arxiv.org/abs/2512.10220
备注：24 pages
摘要：Loog、Mey和Viering在最近的一系列工作中强调了学习曲线单调性的性质，它描述了一种算法，对于给定家族中的任何底层数据分布，它只会在给定更多数据的情况下提高平均性能。我们建立了第一个非平凡的单调性保证的极大似然估计在各种指定的参数设置。对于对数损失的序贯预测，我们显示了单调性（实际上是完全单调性）的正向KL发散高斯向量与未知的协方差和已知或未知的平均值，以及伽玛变量与未知的尺度参数。高斯设置在上述作品中被明确强调为开放的，即使在一维中也是如此。最后，我们观察到，反向KL分歧，一个民间传说的技巧产生单调性非常一般的指数家庭。 本文的所有结果都是由GPT-5.2 Pro的变体推导出来的。人类没有提供任何证明策略或中间论点，只是促使模型继续开发额外的结果，并验证和转录其证明。
摘要：The property of learning-curve monotonicity, highlighted in a recent series of work by Loog, Mey and Viering, describes algorithms which only improve in average performance given more data, for any underlying data distribution within a given family. We establish the first nontrivial monotonicity guarantees for the maximum likelihood estimator in a variety of well-specified parametric settings. For sequential prediction with log loss, we show monotonicity (in fact complete monotonicity) of the forward KL divergence for Gaussian vectors with unknown covariance and either known or unknown mean, as well as for Gamma variables with unknown scale parameter. The Gaussian setting was explicitly highlighted as open in the aforementioned works, even in dimension 1. Finally we observe that for reverse KL divergence, a folklore trick yields monotonicity for very general exponential families. All results in this paper were derived by variants of GPT-5.2 Pro. Humans did not provide any proof strategies or intermediate arguments, but only prompted the model to continue developing additional results, and verified and transcribed its proofs.
【24】A Model-Guided Neural Network Method for the Inverse Scattering Problem
标题：逆散射问题的模型引导神经网络方法
链接：https://arxiv.org/abs/2512.10123
备注：28 pages
摘要：逆介质散射是医学成像、遥感和无损检测中出现的一种不适定的非线性波基成像问题。相对于经典的基于优化的方法，机器学习（ML）方法在捕获成像目标的先验知识方面提供了更快的推理速度和灵活性;然而，它们在散射行为高度非线性的情况下表现不佳。一个关键的限制是，ML方法很难纳入管理散射过程的物理学，这通常是从训练数据中隐含地推断出来的，或者通过架构设计松散地执行。在本文中，我们提出了一种方法，赋予机器学习框架以问题物理学的明确知识，其形式为代表正向模型的可微求解器。所提出的方法逐步完善重建的散射电位使用测量波的频率增加，以下一个经典的策略，以稳定恢复。从经验上讲，我们发现，我们的方法提供了高质量的重建在一小部分的计算或采样成本的竞争方法。
摘要：Inverse medium scattering is an ill-posed, nonlinear wave-based imaging problem arising in medical imaging, remote sensing, and non-destructive testing. Machine learning (ML) methods offer increased inference speed and flexibility in capturing prior knowledge of imaging targets relative to classical optimization-based approaches; however, they perform poorly in regimes where the scattering behavior is highly nonlinear. A key limitation is that ML methods struggle to incorporate the physics governing the scattering process, which are typically inferred implicitly from the training data or loosely enforced via architectural design. In this paper, we present a method that endows a machine learning framework with explicit knowledge of problem physics, in the form of a differentiable solver representing the forward model. The proposed method progressively refines reconstructions of the scattering potential using measurements at increasing wave frequencies, following a classical strategy to stabilize recovery. Empirically, we find that our method provides high-quality reconstructions at a fraction of the computational or sampling costs of competing approaches.
其他(34篇)
【1】Bidirectional Normalizing Flow: From Data to Noise and Back
标题：双向规范化流程：从数据到噪音再回来
链接：https://arxiv.org/abs/2512.10953
备注：Tech report
摘要：规范化流（NF）已经被建立为生成建模的原则框架。标准NF由正向过程和反向过程组成：正向过程将数据映射为噪声，而反向过程通过反转噪声来生成样本。典型的NF正向变换受到显式可逆性的约束，确保反向过程可以作为其精确的解析逆。最近的发展，在TARFlow及其变种振兴NF方法结合Transformers和自回归流，但也暴露了因果解码的一个主要瓶颈。在这项工作中，我们引入双向规范化流（$\textbf{BiFlow}$），一个框架，消除了需要一个精确的解析逆。BiFlow学习近似底层噪声到数据逆映射的逆模型，从而实现更灵活的损失函数和架构。ImageNet上的实验表明，与因果解码对应物相比，BiFlow提高了生成质量，同时将采样速度提高了两个数量级。BiFlow在基于NF的方法中产生最先进的结果，在单评估（“1-NFE”）方法中具有竞争力的性能。在最近关于NF的令人鼓舞的进展之后，我们希望我们的工作将进一步引起人们对这一经典范式的关注。
摘要：Normalizing Flows (NFs) have been established as a principled framework for generative modeling. Standard NFs consist of a forward process and a reverse process: the forward process maps data to noise, while the reverse process generates samples by inverting it. Typical NF forward transformations are constrained by explicit invertibility, ensuring that the reverse process can serve as their exact analytic inverse. Recent developments in TARFlow and its variants have revitalized NF methods by combining Transformers and autoregressive flows, but have also exposed causal decoding as a major bottleneck. In this work, we introduce Bidirectional Normalizing Flow ($\textbf{BiFlow}$), a framework that removes the need for an exact analytic inverse. BiFlow learns a reverse model that approximates the underlying noise-to-data inverse mapping, enabling more flexible loss functions and architectures. Experiments on ImageNet demonstrate that BiFlow, compared to its causal decoding counterpart, improves generation quality while accelerating sampling by up to two orders of magnitude. BiFlow yields state-of-the-art results among NF-based methods and competitive performance among single-evaluation ("1-NFE") methods. Following recent encouraging progress on NFs, we hope our work will draw further attention to this classical paradigm.
【2】Hierarchical Dataset Selection for High-Quality Data Sharing
标题：分层数据集选择以实现高质量数据共享
链接：https://arxiv.org/abs/2512.10952
摘要：现代机器学习的成功取决于获得高质量的训练数据。在许多现实世界的场景中，例如从公共存储库获取数据或跨机构共享数据，数据自然会被组织成离散的数据集，这些数据集在相关性、质量和实用性方面各不相同。因此，选择哪些存储库或机构来搜索有用的数据集，以及将哪些数据集纳入模型训练是关键的决策，但大多数现有方法选择单个样本并将所有数据视为同等相关，忽略了数据集及其来源之间的差异。在这项工作中，我们正式定义了数据集选择的任务：从大型异构池中选择整个数据集，以在资源限制下提高下游性能。我们提出了通过层次结构的数据集选择（DaSH），这是一种数据集选择方法，它对数据集和组的效用进行建模（例如，集合，机构）水平，从而能够从有限的观察中进行有效的概括。在两个公共基准（Digit-Five和DomainNet）中，DaSH在准确性方面优于最先进的数据选择基线，高达26.2%，同时需要的探索步骤明显减少。消融显示DaSH对低资源设置和缺乏相关数据集具有鲁棒性，使其适合于在实际的多源学习工作流中进行可扩展和自适应的数据集选择。
摘要：The success of modern machine learning hinges on access to high-quality training data. In many real-world scenarios, such as acquiring data from public repositories or sharing across institutions, data is naturally organized into discrete datasets that vary in relevance, quality, and utility. Selecting which repositories or institutions to search for useful datasets, and which datasets to incorporate into model training are therefore critical decisions, yet most existing methods select individual samples and treat all data as equally relevant, ignoring differences between datasets and their sources. In this work, we formalize the task of dataset selection: selecting entire datasets from a large, heterogeneous pool to improve downstream performance under resource constraints. We propose Dataset Selection via Hierarchies (DaSH), a dataset selection method that models utility at both dataset and group (e.g., collections, institutions) levels, enabling efficient generalization from limited observations. Across two public benchmarks (Digit-Five and DomainNet), DaSH outperforms state-of-the-art data selection baselines by up to 26.2% in accuracy, while requiring significantly fewer exploration steps. Ablations show DaSH is robust to low-resource settings and lack of relevant datasets, making it suitable for scalable and adaptive dataset selection in practical multi-source learning workflows.
【3】Any4D: Unified Feed-Forward Metric 4D Reconstruction
标题：Any4D：统一的前向度量4D重建
链接：https://arxiv.org/abs/2512.10935
备注：Project Website: https://any-4d.github.io/
摘要：我们提出了Any 4D，一个可扩展的多视图Transformer，用于度量尺度，密集前馈4D重建。Any 4D直接生成N帧的每像素运动和几何预测，与通常专注于2视图密集场景流或稀疏3D点跟踪的先前工作相反。此外，与最近从单目RGB视频进行4D重建的其他方法不同，Any 4D可以处理额外的模态和传感器，例如RGB-D帧，基于IMU的自运动和雷达多普勒测量。允许这种灵活框架的关键创新之一是4D场景的模块化表示;具体地说，每视图4D预测使用在局部相机坐标中表示的各种自我中心因素（深度图和相机固有）和在全局世界坐标中表示的allocentric因素（相机外部和场景流）进行编码。我们在不同的设置中实现了卓越的性能-无论是在准确性（2- 3倍的误差降低）还是计算效率（15倍的速度）方面，都为多个下游应用开辟了道路。
摘要 ：We present Any4D, a scalable multi-view transformer for metric-scale, dense feed-forward 4D reconstruction. Any4D directly generates per-pixel motion and geometry predictions for N frames, in contrast to prior work that typically focuses on either 2-view dense scene flow or sparse 3D point tracking. Moreover, unlike other recent methods for 4D reconstruction from monocular RGB videos, Any4D can process additional modalities and sensors such as RGB-D frames, IMU-based egomotion, and Radar Doppler measurements, when available. One of the key innovations that allows for such a flexible framework is a modular representation of a 4D scene; specifically, per-view 4D predictions are encoded using a variety of egocentric factors (depthmaps and camera intrinsics) represented in local camera coordinates, and allocentric factors (camera extrinsics and scene flow) represented in global world coordinates. We achieve superior performance across diverse setups - both in terms of accuracy (2-3X lower error) and compute efficiency (15X faster), opening avenues for multiple downstream applications.
【4】Decoupled Q-Chunking
标题：解耦Q-组块
链接：https://arxiv.org/abs/2512.10926
备注：76 pages, 14 figures
摘要：时间差分（TD）方法通过从它们自己的未来值预测中自举来有效地学习状态和动作值，但是这样的自举机制易于产生自举偏差，其中值目标中的误差跨步骤累积并导致有偏差的值估计。最近的研究提出使用分块批评，它估计短动作序列（“块”）而不是单个动作的价值，加速价值备份。然而，从分块的批评者中提取政策是具有挑战性的：政策必须输出整个动作块开环，这对于需要政策反应性的环境来说可能是次优的，并且特别是当块长度增加时，建模也具有挑战性。我们的关键见解是将批评者的块长度与策略的块长度解耦，允许策略在较短的动作块上操作。我们提出了一种新的算法，实现了这一点，通过优化政策对蒸馏的批评部分动作块，构造乐观备份从原来的分块的批评，以近似的最大值时，部分动作块扩展到一个完整的。这种设计保留了多步值传播的优点，同时避开了开环次优性和学习长动作块的动作块策略的困难。我们评估我们的方法具有挑战性的，长期离线目标条件任务，并表明它可靠地优于以前的方法。代码：github.com/ColinQiyangLi/dqc.
摘要：Temporal-difference (TD) methods learn state and action values efficiently by bootstrapping from their own future value predictions, but such a self-bootstrapping mechanism is prone to bootstrapping bias, where the errors in the value targets accumulate across steps and result in biased value estimates. Recent work has proposed to use chunked critics, which estimate the value of short action sequences ("chunks") rather than individual actions, speeding up value backup. However, extracting policies from chunked critics is challenging: policies must output the entire action chunk open-loop, which can be sub-optimal for environments that require policy reactivity and also challenging to model especially when the chunk length grows. Our key insight is to decouple the chunk length of the critic from that of the policy, allowing the policy to operate over shorter action chunks. We propose a novel algorithm that achieves this by optimizing the policy against a distilled critic for partial action chunks, constructed by optimistically backing up from the original chunked critic to approximate the maximum value achievable when a partial action chunk is extended to a complete one. This design retains the benefits of multi-step value propagation while sidestepping both the open-loop sub-optimality and the difficulty of learning action chunking policies for long action chunks. We evaluate our method on challenging, long-horizon offline goal-conditioned tasks and show that it reliably outperforms prior methods. Code: github.com/ColinQiyangLi/dqc.
【5】A Differentiable Digital Twin of Distributed Link Scheduling for Contention-Aware Networking
标题：竞争感知网络的分布式链路调度的差异数字孪生
链接：https://arxiv.org/abs/2512.10874
备注：5 pages, 8 figures, presented in Asilomar Conference on Signals, Systems, and Computers 2025
摘要：许多有线网络中的路由和流量优化问题可以有效地解决使用最小费用流公式。然而，这种方法并没有扩展到无线多跳网络，其中固定链路容量和线性成本结构的假设由于共享频谱资源的竞争而崩溃。关键的挑战是无线链路的长期容量成为其网络环境的非线性函数，包括网络拓扑、链路质量和分配给相邻链路的流量。在这项工作中，我们追求一个新的方向，随机介质访问控制下的无线网络建模，通过开发一个分析网络数字孪生（NDT），预测链路占空比从网络环境。我们广义的随机竞争，找到一个最大独立集（MIS）的冲突图使用加权卢比的算法，推导出一个分析模型的链接占空比，并引入一个迭代过程，解决了循环之间的依赖占空比，链路容量，和竞争概率。我们的数值实验表明，建议NDT准确地预测链路占空比和拥塞模式，高达5000倍的加速比超过数据包级仿真，并使我们能够优化链路调度使用梯度下降减少拥塞和无线电足迹。
摘要：Many routing and flow optimization problems in wired networks can be solved efficiently using minimum cost flow formulations. However, this approach does not extend to wireless multi-hop networks, where the assumptions of fixed link capacity and linear cost structure collapse due to contention for shared spectrum resources. The key challenge is that the long-term capacity of a wireless link becomes a non-linear function of its network context, including network topology, link quality, and the traffic assigned to neighboring links. In this work, we pursue a new direction of modeling wireless network under randomized medium access control by developing an analytical network digital twin (NDT) that predicts link duty cycles from network context. We generalize randomized contention as finding a Maximal Independent Set (MIS) on the conflict graph using weighted Luby's algorithm, derive an analytical model of link duty cycles, and introduce an iterative procedure that resolves the circular dependency among duty cycle, link capacity, and contention probability. Our numerical experiments show that the proposed NDT accurately predicts link duty cycles and congestion patterns with up to a 5000x speedup over packet-level simulation, and enables us to optimize link scheduling using gradient descent for reduced congestion and radio footprint.
【6】Interpretable and Steerable Concept Bottleneck Sparse Autoencoders
标题：可解释和可控制的概念瓶颈稀疏自动编码器
链接：https://arxiv.org/abs/2512.10805
摘要：稀疏自动编码器（SAE）承诺在LLM和LVLM中实现机械可解释性，概念发现和模型转向的统一方法。然而，实现这种潜力需要学习的功能是可解释和可操纵的。为此，我们引入了两个新的计算成本低廉的可解释性和可操纵性指标，并进行了系统的分析LVLM。我们的分析揭示了两个观察结果;（i）大多数SAE神经元表现出低可解释性或低可操纵性或两者兼而有之，使它们对下游使用无效;（ii）由于SAE的无监督性质，用户期望的概念通常不存在于学习的字典中，从而限制了它们的实际效用。为了解决这些限制，我们提出了概念瓶颈稀疏自动编码器（CB-SAE）-一种新的事后框架，修剪低效用神经元，并增加了潜在的空间与一个轻量级的概念瓶颈对准用户定义的概念集。由此产生的CB-SAE在LVLM和图像生成任务中将可解释性提高了+32.1%，可操纵性提高了+14.5%。我们将提供我们的代码和模型权重。
摘要：Sparse autoencoders (SAEs) promise a unified approach for mechanistic interpretability, concept discovery, and model steering in LLMs and LVLMs. However, realizing this potential requires that the learned features be both interpretable and steerable. To that end, we introduce two new computationally inexpensive interpretability and steerability metrics and conduct a systematic analysis on LVLMs. Our analysis uncovers two observations; (i) a majority of SAE neurons exhibit either low interpretability or low steerability or both, rendering them ineffective for downstream use; and (ii) due to the unsupervised nature of SAEs, user-desired concepts are often absent in the learned dictionary, thus limiting their practical utility. To address these limitations, we propose Concept Bottleneck Sparse Autoencoders (CB-SAE) - a novel post-hoc framework that prunes low-utility neurons and augments the latent space with a lightweight concept bottleneck aligned to a user-defined concept set. The resulting CB-SAE improves interpretability by +32.1% and steerability by +14.5% across LVLMs and image generation tasks. We will make our code and model weights available.
【7】OPV: Outcome-based Process Verifier for Efficient Long Chain-of-Thought Verification
标题：OPV：基于结果的流程验证器，用于有效的长思想链验证
链接：https://arxiv.org/abs/2512.10756
摘要：大型语言模型（LLM）通过具有可验证奖励的强化学习（RLVR）在解决复杂推理任务方面取得了重大进展。这一进步也与可靠的核查人员的自动监督密不可分。然而，目前基于结果的验证器（OV）无法检查长推理思想链（CoT）中不可靠的中间步骤。与此同时，当前基于过程的验证器（PV）在可靠地检测复杂的长CoT中的错误方面存在困难，这受到由于人类注释的高昂成本而导致的高质量注释稀缺的限制。因此，我们提出了基于结果的过程验证器（OPV），它验证了从长CoT中总结结果的合理过程，以实现准确和高效的验证，并实现大规模注释。为了使拟议的验证，我们采用了一个迭代的主动学习框架与专家注释，以逐步提高验证能力的OPV更少的注释成本。具体地，在每次迭代中，当前最佳OPV的最不确定的情况被注释，然后随后用于通过拒绝微调（RFT）和RLVR为下一轮训练新的OPV。大量的实验证明了OPV的优越性能和广泛的适用性。它在我们的OPV-Bench上取得了最先进的结果，表现优于更大的开源模型，如Qwen 3-Max-Preview，F1得分为83.1，而不是76.3。此外，OPV有效地检测合成数据集中的假阳性，与专家评估密切相关。当与策略模型协作时，OPV始终如一地产生性能增益，例如，随着计算预算的增加，DeepSeek-R1-Distill-Qwen-32 B在AIME 2025上的准确率从55.2%提高到73.3%。
摘要：Large language models (LLMs) have achieved significant progress in solving complex reasoning tasks by Reinforcement Learning with Verifiable Rewards (RLVR). This advancement is also inseparable from the oversight automated by reliable verifiers. However, current outcome-based verifiers (OVs) are unable to inspect the unreliable intermediate steps in the long reasoning chains of thought (CoTs). Meanwhile, current process-based verifiers (PVs) have difficulties in reliably detecting errors in the complex long CoTs, limited by the scarcity of high-quality annotations due to the prohibitive costs of human annotations. Therefore, we propose the Outcome-based Process Verifier (OPV), which verifies the rationale process of summarized outcomes from long CoTs to achieve both accurate and efficient verification and enable large-scale annotation. To empower the proposed verifier, we adopt an iterative active learning framework with expert annotations to progressively improve the verification capability of OPV with fewer annotation costs. Specifically, in each iteration, the most uncertain cases of the current best OPV are annotated and then subsequently used to train a new OPV through Rejection Fine-Tuning (RFT) and RLVR for the next round. Extensive experiments demonstrate OPV's superior performance and broad applicability. It achieves new state-of-the-art results on our held-out OPV-Bench, outperforming much larger open-source models such as Qwen3-Max-Preview with an F1 score of 83.1 compared to 76.3. Furthermore, OPV effectively detects false positives within synthetic dataset, closely align with expert assessment. When collaborating with policy models, OPV consistently yields performance gains, e.g., raising the accuracy of DeepSeek-R1-Distill-Qwen-32B from 55.2% to 73.3% on AIME2025 as the compute budget scales.
【8】Generalized Spherical Neural Operators: Green's Function Formulation
标题：广义球形神经运算符：格林函数公式
链接：https://arxiv.org/abs/2512.10723
摘要：神经算子为求解参数偏微分方程提供了强大的方法，但将它们扩展到球面域仍然具有挑战性，因为需要保留固有几何形状，同时避免破坏旋转一致性的扭曲。现有的球形算子依赖于旋转等方差，但往往缺乏现实世界的复杂性的灵活性。我们提出了一个通用的运营商设计框架的基础上可设计的球形格林函数及其谐波扩展，建立了坚实的运营商理论基础的球形学习。在此基础上，我们提出了一个绝对和相对位置相关的格林函数，使灵活的平衡等变性和不变性的真实世界的建模。由此产生的算子，格林函数球面神经算子（GSNO）与一种新的频谱学习方法，可以适应各向异性，约束丰富的系统，同时保持频谱效率。为了利用GSNO，我们开发了GSHNet，这是一种分层架构，将多尺度光谱建模与球形上下采样相结合，增强了全局特征表示。在扩散磁共振成像、浅水动力学和全球天气预报方面，GSNO和GSHNet的评价始终优于最先进的方法。我们的研究结果将GSNO定位为球形算子学习的原则性和通用框架，将严格的理论与现实世界的复杂性联系起来。
摘要：Neural operators offer powerful approaches for solving parametric partial differential equations, but extending them to spherical domains remains challenging due to the need to preserve intrinsic geometry while avoiding distortions that break rotational consistency. Existing spherical operators rely on rotational equivariance but often lack the flexibility for real-world complexity. We propose a general operator-design framework based on the designable spherical Green's function and its harmonic expansion, establishing a solid operator-theoretic foundation for spherical learning. Based on this, we propose an absolute and relative position-dependent Green's function that enables flexible balance of equivariance and invariance for real-world modeling. The resulting operator, Green's-function Spherical Neural Operator (GSNO) with a novel spectral learning method, can adapt to anisotropic, constraint-rich systems while retaining spectral efficiency. To exploit GSNO, we develop GSHNet, a hierarchical architecture that combines multi-scale spectral modeling with spherical up-down sampling, enhancing global feature representation. Evaluations on diffusion MRI, shallow water dynamics, and global weather forecasting, GSNO and GSHNet consistently outperform state-of-the-art methods. Our results position GSNO as a principled and general framework for spherical operator learning, bridging rigorous theory with real-world complexity.
【9】Sharp Monocular View Synthesis in Less Than a Second
标题：不到一秒的清晰单目视图合成
链接：https://arxiv.org/abs/2512.10685
备注：Code and weights available at https://github.com/apple/ml-sharp
摘要：我们提出了SHARP，一种从单个图像合成真实感视图的方法。给定一张照片，SHARP回归所描绘场景的3D高斯表示的参数。这是在标准GPU上通过神经网络的单个前馈传递在不到一秒的时间内完成的。由SHARP产生的3D高斯表示可以实时渲染，为附近的视图产生高分辨率的照片级真实感图像。表示是公制的，具有绝对比例，支持公制相机移动。实验结果表明，SHARP提供了强大的跨数据集的zero-shot泛化。它在多个数据集上设定了一个新的技术水平，与最佳先验模型相比，LPIPS减少了25-34%，DISTS减少了21-43%，同时将合成时间降低了三个数量级。代码和重量请访问https://github.com/apple/ml-sharp
摘要：We present SHARP, an approach to photorealistic view synthesis from a single image. Given a single photograph, SHARP regresses the parameters of a 3D Gaussian representation of the depicted scene. This is done in less than a second on a standard GPU via a single feedforward pass through a neural network. The 3D Gaussian representation produced by SHARP can then be rendered in real time, yielding high-resolution photorealistic images for nearby views. The representation is metric, with absolute scale, supporting metric camera movements. Experimental results demonstrate that SHARP delivers robust zero-shot generalization across datasets. It sets a new state of the art on multiple datasets, reducing LPIPS by 25-34% and DISTS by 21-43% versus the best prior model, while lowering the synthesis time by three orders of magnitude. Code and weights are provided at https://github.com/apple/ml-sharp
【10】Evaluating Gemini Robotics Policies in a Veo World Simulator
标题：在Veo世界模拟器中评估Gemini机器人政策
链接：https://arxiv.org/abs/2512.10675
摘要 ：生成世界模型在不同环境中模拟与可视化策略的交互方面具有巨大的潜力。前沿视频模型可以以可扩展和通用的方式生成逼真的观察和环境交互。然而，视频模型在机器人技术中的使用主要限于分发评估，即，类似于用于训练策略或微调基本视频模型的场景。在本报告中，我们证明了视频模型可用于机器人中的整个策略评估用例：从评估标称性能到分布外（OOD）泛化，以及探测物理和语义安全性。我们介绍了一个生成的评价系统建立在一个前沿的视频基础模型（Veo）。该系统进行了优化，以支持机器人动作调节和多视图一致性，同时集成生成图像编辑和多视图完成，以合成现实世界场景的现实变化沿多个轴的泛化。我们证明，该系统保留了视频模型的基本功能，使已编辑的场景，包括新的交互对象，新的视觉背景，和新的干扰对象的准确模拟。这种保真度使得能够准确地预测不同策略在标称和OOD条件下的相对性能，确定不同泛化轴对策略性能的相对影响，并执行策略的红色分组以暴露违反物理或语义安全约束的行为。我们通过对8个Gemini Robotics策略检查点和5个双手操纵器任务的1600多个真实世界评估来验证这些功能。
摘要：Generative world models hold significant potential for simulating interactions with visuomotor policies in varied environments. Frontier video models can enable generation of realistic observations and environment interactions in a scalable and general manner. However, the use of video models in robotics has been limited primarily to in-distribution evaluations, i.e., scenarios that are similar to ones used to train the policy or fine-tune the base video model. In this report, we demonstrate that video models can be used for the entire spectrum of policy evaluation use cases in robotics: from assessing nominal performance to out-of-distribution (OOD) generalization, and probing physical and semantic safety. We introduce a generative evaluation system built upon a frontier video foundation model (Veo). The system is optimized to support robot action conditioning and multi-view consistency, while integrating generative image-editing and multi-view completion to synthesize realistic variations of real-world scenes along multiple axes of generalization. We demonstrate that the system preserves the base capabilities of the video model to enable accurate simulation of scenes that have been edited to include novel interaction objects, novel visual backgrounds, and novel distractor objects. This fidelity enables accurately predicting the relative performance of different policies in both nominal and OOD conditions, determining the relative impact of different axes of generalization on policy performance, and performing red teaming of policies to expose behaviors that violate physical or semantic safety constraints. We validate these capabilities through 1600+ real-world evaluations of eight Gemini Robotics policy checkpoints and five tasks for a bimanual manipulator.
【11】DCFO Additional Material
标题：DCFO附加材料
链接：https://arxiv.org/abs/2512.10659
摘要：离群值检测识别显著偏离大多数数据分布的数据点。解释异常值对于理解有助于其检测的潜在因素，验证其重要性以及识别潜在的偏差或错误至关重要。有效的解释提供了可操作的见解，有助于采取预防措施，避免未来出现类似的异常值。反事实解释通过确定改变其预测所需的最小变化来澄清为什么特定数据点被归类为离群值。尽管有价值，但大多数现有的反事实解释方法忽视了离群点检测所带来的独特挑战，并且未能针对经典的、广泛采用的离群点检测算法。LOF（Local Outlier Factor）是最流行的无监督离群点检测方法之一，通过相对局部密度来量化离群点。尽管LOF在不同的应用程序中广泛使用，但它缺乏可解释性。为了解决这个问题，我们引入了基于密度的异常值反事实（DCFO），这是一种专门为LOF生成反事实解释的新方法。DCFO将数据空间划分为LOF表现平滑的区域，从而实现高效的基于梯度的优化。在50个OpenML数据集上进行的广泛实验验证表明，DCFO始终优于基准竞争对手，提供了生成的反事实的卓越接近性和有效性。
摘要：Outlier detection identifies data points that significantly deviate from the majority of the data distribution. Explaining outliers is crucial for understanding the underlying factors that contribute to their detection, validating their significance, and identifying potential biases or errors. Effective explanations provide actionable insights, facilitating preventive measures to avoid similar outliers in the future. Counterfactual explanations clarify why specific data points are classified as outliers by identifying minimal changes required to alter their prediction. Although valuable, most existing counterfactual explanation methods overlook the unique challenges posed by outlier detection, and fail to target classical, widely adopted outlier detection algorithms. Local Outlier Factor (LOF) is one the most popular unsupervised outlier detection methods, quantifying outlierness through relative local density. Despite LOF's widespread use across diverse applications, it lacks interpretability. To address this limitation, we introduce Density-based Counterfactuals for Outliers (DCFO), a novel method specifically designed to generate counterfactual explanations for LOF. DCFO partitions the data space into regions where LOF behaves smoothly, enabling efficient gradient-based optimisation. Extensive experimental validation on 50 OpenML datasets demonstrates that DCFO consistently outperforms benchmarked competitors, offering superior proximity and validity of generated counterfactuals.
【12】Token Sample Complexity of Attention
标题：代币样本注意力的复杂性
链接：https://arxiv.org/abs/2512.10656
摘要：随着大型语言模型中的上下文窗口不断扩大，描述注意力在极端序列长度下的行为是至关重要的。我们引入令牌样本复杂度：在$n$令牌计算的注意力收敛到其无限令牌限制的速度。我们估计有限-$n$收敛界在两个层面上：逐点一致收敛的注意力地图，和收敛的时刻变换令牌分布。对于紧支撑的（更一般的亚高斯）分布，我们的第一个结果表明，注意力地图以$C（R）/\sqrt{n}$的速率均匀地收敛于半径为$R$的球上，其中$C（R）$随着$R$呈指数增长。对于较大的$R$，这个估计失去了实际价值，我们的第二个结果通过建立转换后的分布（注意力层的令牌输出）的矩的收敛率来解决这个问题。在这种情况下，比率是$C '（R）/n^β$，其中$β
摘要：As context windows in large language models continue to expand, it is essential to characterize how attention behaves at extreme sequence lengths. We introduce token-sample complexity: the rate at which attention computed on $n$ tokens converges to its infinite-token limit. We estimate finite-$n$ convergence bounds at two levels: pointwise uniform convergence of the attention map, and convergence of moments for the transformed token distribution. For compactly supported (and more generally sub-Gaussian) distributions, our first result shows that the attention map converges uniformly on a ball of radius $R$ at rate $C(R)/\sqrt{n}$, where $C(R)$ grows exponentially with $R$. For large $R$, this estimate loses practical value, and our second result addresses this issue by establishing convergence rates for the moments of the transformed distribution (the token output of the attention layer). In this case, the rate is $C'(R)/n^β$ with $β
【13】Supporting Migration Policies with Forecasts: Illegal Border Crossings in Europe through a Mixed Approach
标题：通过预测支持移民政策：通过混合方法解决欧洲非法越境问题
链接：https://arxiv.org/abs/2512.10633
备注：17 pages, 6 figures, 2 tables + supplementary material with 20 pages, 21 figures, 2 tables
摘要：本文提出了一种混合方法来预测欧洲五条主要移民路线的非法越境情况，时间跨度为一年。该方法将机器学习技术与移民专家的定性见解相结合。这种方法旨在通过纳入人类评估的协变量来提高数据驱动模型的预测能力，这是一种创新，可以解决移民模式突然变化和传统数据集局限性带来的挑战。拟议的方法直接响应欧盟《移民和庇护公约》中概述的预测需求，支持《庇护和移民管理条例》。它旨在提供与政策相关的预测，为欧盟成员国的战略决策、预警系统和团结机制提供信息。通过将数据驱动的建模与专家判断相结合，这项工作与现有的学术建议保持一致，并引入了为欧盟移民治理量身定制的新型操作工具。该方法进行了测试和验证与已知的数据，以证明其适用性和可靠性的迁移相关的政策背景下。
摘要 ：This paper presents a mixed-methodology to forecast illegal border crossings in Europe across five key migratory routes, with a one-year time horizon. The methodology integrates machine learning techniques with qualitative insights from migration experts. This approach aims at improving the predictive capacity of data-driven models through the inclusion of a human-assessed covariate, an innovation that addresses challenges posed by sudden shifts in migration patterns and limitations in traditional datasets. The proposed methodology responds directly to the forecasting needs outlined in the EU Pact on Migration and Asylum, supporting the Asylum and Migration Management Regulation (AMMR). It is designed to provide policy-relevant forecasts that inform strategic decisions, early warning systems, and solidarity mechanisms among EU Member States. By joining data-driven modeling with expert judgment, this work aligns with existing academic recommendations and introduces a novel operational tool tailored for EU migration governance. The methodology is tested and validated with known data to demonstrate its applicability and reliability in migration-related policy context.
【14】Confucius Code Agent: An Open-sourced AI Software Engineer at Industrial Scale
标题：孔子代码代理：工业规模的开源人工智能软件工程师
链接：https://arxiv.org/abs/2512.10398
摘要：现实世界的人工智能软件工程要求编码代理能够在大量存储库中进行推理，在长时间会话中保持持久的内存，并在测试时稳健地协调复杂的工具链。现有的开源编码代理提供了透明度，但在推向这些工业规模的工作负载时经常达不到要求，而专有编码代理提供了强大的实用性能，但可扩展性，可解释性和可控性有限。我们介绍了孔子代码代理（CCA），这是一个开源的人工智能软件工程师，可以在工业规模上运行。CCA建立在孔子SDK之上，孔子SDK是一个开源的代理开发平台，围绕三个互补的角度设计：代理体验（AX），用户体验（UX）和开发人员体验（DX）。SDK引入了一个统一的协调器，具有用于长上下文推理的分层工作记忆，用于跨会话持续学习的持久笔记系统，以及用于强大工具使用的模块化扩展模块。此外，元代理通过构建-测试-改进循环自动化代理配置的合成，评估和细化，从而实现新任务，环境和工具堆栈的快速代理开发。通过这些机制在孔子SDK上实例化，CCA在现实世界的软件工程任务上提供了强大的性能。在SWE-Bench-Pro上，CCA实现了54.3%的最先进的Resolve@1性能，大大提高了以前的编码代理。Confucius SDK和CCA共同为人工智能代理提供了透明、可扩展和可复制的基础，弥合了研究原型和生产级系统之间的差距，并支持工业规模的代理开发和部署。
摘要：Real-world AI software engineering demands coding agents that can reason over massive repositories, maintain durable memory across and within long sessions, and robustly coordinate complex toolchains at test time. Existing open-source coding agents provide transparency but frequently fall short when pushed to these industrial-scale workloads, while proprietary coding agents offer strong practical performance but limited extensibility, interpretability, and controllability. We present the Confucius Code Agent (CCA), an open-sourced AI software engineer that can operate at an industrial scale. CCA is built atop the Confucius SDK, an open-sourced agent development platform designed around three complementary perspectives: Agent Experience (AX), User Experience (UX), and Developer Experience (DX). The SDK introduces a unified orchestrator with hierarchical working memory for long-context reasoning, a persistent note-taking system for cross-session continual learning, and a modular extension module for robust tool use. Moreover, a meta-agent automates the synthesis, evaluation, and refinement of agent configurations through a build-test-improve loop, enabling rapid agent development on new tasks, environments, and tool stacks. Instantiated on Confucius SDK with these mechanisms, CCA delivers strong performance on real-world software engineering tasks. On SWE-Bench-Pro, CCA achieves a state-of-the-art Resolve@1 performance of 54.3%, substantially improving over prior coding agents. Together, the Confucius SDK and CCA provide a transparent, extensible, and reproducible foundation for AI agents, bridge gaps between research prototypes and production-grade systems, and support agent development and deployment at industrial scale.
【15】Fitting magnetization data using continued fraction of straight lines
标题：使用直线的连续分数来匹配磁性数据
链接：https://arxiv.org/abs/2512.10390
备注：17 pages, 12 figures, 4 tables
摘要：铁磁物质响应于外部施加的磁场的磁化强度随着场的强度而增加。这是因为在微观水平上，物质的某些区域或域中的磁矩越来越多地与所施加的场对准，而未对准的域的量减少。这种磁畴与所施加的磁场的对准形成磁化的非线性的物理基础。本文将非线性函数近似为直线连分式的组合。由此产生的拟合被用来解释在增长和收缩磁畴的非线性行为。这里使用的直线的连分数是一个代数表达式，可用于使用非线性回归估计参数。
摘要：Magnetization of a ferromagnetic substance in response to an externally applied magnetic field increases with the strength of the field. This is because at the microscopic level, magnetic moments in certain regions or domains of the substance increasingly align with the applied field, while the amount of misaligned domains decreases. The alignment of such magnetic domains with an applied magnetic field forms the physical basis for the nonlinearity of magnetization. In this paper, the nonlinear function is approximated as a combination of continued fraction of straight lines. The resulting fit is used to interpret the nonlinear behavior in both growing and shrinking magnetic domains. The continued fraction of straight lines used here is an algebraic expression which can be used to estimate parameters using nonlinear regression.
【16】An Interpretable AI Tool for SAVR vs TAVR in Low to Intermediate Risk Patients with Severe Aortic Stenosis
标题：用于严重主动脉狭窄的低至中度风险患者的SAVR与TAVR的可解释人工智能工具
链接：https://arxiv.org/abs/2512.10308
摘要：背景在临床实践中，受患者异质性和机构偏好的驱动，手术（SAVR）和经导管（TAVR）主动脉瓣置换术之间重度主动脉瓣狭窄低至中等风险患者的治疗选择仍然存在差异。虽然现有的模型预测术后风险，但缺乏可解释的，个性化的治疗建议，直接优化长期结果。 方法.我们引入了一个可解释的规范性框架，该框架集成了预后匹配、反事实结局建模和最优策略树（OPT），以推荐最小化预期5年死亡率的治疗方案。使用来自Hartford医院和圣文森特医院的数据，我们通过预后匹配和样本加权模拟随机化，并估计SAVR和TAVR下的反事实死亡率。在这些反事实预测的基础上训练的政策模型将患者划分为临床一致的亚组，并规定与较低估计风险相关的治疗。 调查结果。如果应用OPT处方，反事实评估显示，与现实生活处方相比，Hartford和St. Vincent的5年死亡率估计分别降低20.3%和13.8%，显示出对来自不同机构的未知数据的有希望的普遍性。学习的决策边界与现实世界的结果和临床观察一致。 口译。据我们所知，我们的可解释性规定性框架是第一个为TAVR与SAVR提供透明，数据驱动的建议，可改善内部和外部队列的估计长期结局，同时保持临床基础，并有助于结构性心脏病精准医学的更系统和循证方法。
摘要：Background. Treatment selection for low to intermediate risk patients with severe aortic stenosis between surgical (SAVR) and transcatheter (TAVR) aortic valve replacement remains variable in clinical practice, driven by patient heterogeneity and institutional preferences. While existing models predict postprocedural risk, there is a lack of interpretable, individualized treatment recommendations that directly optimize long-term outcomes. Methods. We introduce an interpretable prescriptive framework that integrates prognostic matching, counterfactual outcome modeling, and an Optimal Policy Tree (OPT) to recommend the treatment minimizing expected 5-year mortality. Using data from Hartford Hospital and St. Vincent's Hospital, we emulate randomization via prognostic matching and sample weighting and estimate counterfactual mortality under both SAVR and TAVR. The policy model, trained on these counterfactual predictions, partitions patients into clinically coherent subgroups and prescribes the treatment associated with lower estimated risk. Findings. If the OPT prescriptions are applied, counterfactual evaluation showed an estimated reduction in 5-year mortality of 20.3\% in Hartford and 13.8\% in St. Vincent's relative to real-life prescriptions, showing promising generalizability to unseen data from a different institution. The learned decision boundaries aligned with real-world outcomes and clinical observations. Interpretation. Our interpretable prescriptive framework is, to the best of our knowledge, the first to provide transparent, data-driven recommendations for TAVR versus SAVR that improve estimated long-term outcomes both in an internal and external cohort, while remaining clinically grounded and contributing toward a more systematic and evidence-based approach to precision medicine in structural heart disease.
【17】Design Space Exploration of DMA based Finer-Grain Compute Communication Overlap
标题：基于DMA的细粒度计算通信重叠的设计空间探索
链接：https://arxiv.org/abs/2512.10236
摘要：随着ML训练和推理越来越分布式，通常会部署并行化技术，在分布式系统的GPU上分割ML模型。通过这些技术，依赖于数据的通信和计算操作非常普遍，其中通信是公开的，留下高达1.7倍的理想性能。之前的工作利用了ML模型状态和输入已经分片的事实，并采用了单独计算/通信分片的仔细重叠。虽然这种粗粒度的重叠是有希望的，但在这项工作中，我们提出了一种更细粒度的计算通信重叠，我们称之为FiCCO，我们主张更细粒度，比分片级别更深的重叠，以解锁更广泛的网络拓扑，更细粒度的计算/通信重叠等等。我们表明，FiCCO开辟了一个更广泛的设计空间的执行时间表比可能单独在分片级别。同时，将ML操作分解为更小的操作（在基于分片的技术和更细粒度的技术中完成）会导致操作级别的低效率损失。为了平衡这两者，我们首先提出了这些低效率损失的详细特征，然后提出了一个设计空间的FiCCO时间表，最后覆盖的时间表与伴随的低效率签名。这样做可以帮助我们设计框架和运行时可以利用的算法，以根据底层ML操作的性质选择定制的FiCCO时间表。最后，为了进一步减少操作重叠所固有的争用低效，我们将通信卸载到GPU DMA引擎。我们从现实的ML部署中评估了几种场景，并证明我们提出的定制时间表可以提供高达1.6倍的加速，并且我们的算法在81%的未知场景中提供了准确的指导。
摘要：As both ML training and inference are increasingly distributed, parallelization techniques that shard (divide) ML model across GPUs of a distributed system, are often deployed. With such techniques, there is a high prevalence of data-dependent communication and computation operations where communication is exposed, leaving as high as 1.7x ideal performance on the table. Prior works harness the fact that ML model state and inputs are already sharded, and employ careful overlap of individual computation/communication shards. While such coarse-grain overlap is promising, in this work, we instead make a case for finer-grain compute-communication overlap which we term FiCCO, where we argue for finer-granularity, one-level deeper overlap than at shard-level, to unlock compute/communication overlap for a wider set of network topologies, finer-grain dataflow and more. We show that FiCCO opens up a wider design space of execution schedules than possible at shard-level alone. At the same time, decomposition of ML operations into smaller operations (done in both shard-based and finer-grain techniques) causes operation-level inefficiency losses. To balance the two, we first present a detailed characterization of these inefficiency losses, then present a design space of FiCCO schedules, and finally overlay the schedules with concomitant inefficiency signatures. Doing so helps us design heuristics that frameworks and runtimes can harness to select bespoke FiCCO schedules based on the nature of underlying ML operations. Finally, to further minimize contention inefficiencies inherent with operation overlap, we offload communication to GPU DMA engines. We evaluate several scenarios from realistic ML deployments and demonstrate that our proposed bespoke schedules deliver up to 1.6x speedup and our heuristics provide accurate guidance in 81% of unseen scenarios.
【18】Federated Domain Generalization with Latent Space Inversion
标题：带潜在空间倒置的联邦域推广
链接：https://arxiv.org/abs/2512.10224
备注：Accepted at ICDM 2025
摘要：联邦域泛化（FedDG）解决了联邦学习框架中客户端之间的分布变化。FedDG方法聚合本地训练的客户端模型的参数，以形成一个全局模型，该模型可以推广到看不见的客户端，同时保护数据隐私。在提高全局模型泛化能力的同时，FedDG中的许多现有方法通过在它们之间共享客户端数据的统计信息来危害隐私。我们的解决方案通过提供新的方法来执行本地客户端培训和模型聚合来解决这个问题。为了改善本地客户端训练，我们在一种新技术的帮助下，在本地模型中执行（域）不变性，这种新技术可以实现更好的客户端隐私。当客户端不\n {i.i.d}时，聚合其本地模型可能会丢弃某些本地适配。为了克服这一点，我们提出了一个\textbf{important weight}聚合策略，以优先考虑在聚合过程中显着影响局部模型预测的参数。我们广泛的实验表明，我们的方法取得了优越的结果比国家的最先进的方法，更少的通信开销。
摘要：Federated domain generalization (FedDG) addresses distribution shifts among clients in a federated learning framework. FedDG methods aggregate the parameters of locally trained client models to form a global model that generalizes to unseen clients while preserving data privacy. While improving the generalization capability of the global model, many existing approaches in FedDG jeopardize privacy by sharing statistics of client data between themselves. Our solution addresses this problem by contributing new ways to perform local client training and model aggregation. To improve local client training, we enforce (domain) invariance across local models with the help of a novel technique, \textbf{latent space inversion}, which enables better client privacy. When clients are not \emph{i.i.d}, aggregating their local models may discard certain local adaptations. To overcome this, we propose an \textbf{important weight} aggregation strategy to prioritize parameters that significantly influence predictions of local models during aggregation. Our extensive experiments show that our approach achieves superior results over state-of-the-art methods with less communication overhead.
【19】Exact Recovery of Non-Random Missing Multidimensional Time Series via Temporal Isometric Delay-Embedding Transform
标题：通过时间等距延迟嵌入变换精确恢复非随机缺失多维时间序列
链接：https://arxiv.org/abs/2512.10191
摘要：非随机缺失数据是多维时间序列中普遍存在但未得到充分处理的缺陷，从根本上威胁到数据驱动分析和决策的可靠性。纯低秩张量补全作为一种经典的数据恢复方法，在处理非随机丢失时，无论在方法上还是在理论上都存在不足。Hankel结构的张量完备化模型为恢复具有非随机缺失模式的多维时间序列提供了一种可行的方法。然而，大多数基于Hankel的多维数据恢复方法都遭受Hankel张量低秩的来源不清楚，并且缺乏对非随机缺失数据的精确恢复理论。为了解决这些问题，我们提出了时间等距延迟嵌入变换，它构造了一个汉克尔张量，其低秩自然是由底层时间序列的平滑性和周期性引起的。利用这一特性，我们开发了\textit{低秩张量完成与时间等距延迟嵌入变换}（LRTC-TIDT）模型，该模型在\textit{张量奇异值分解}（t-SVD）框架下表征了低秩结构。一旦满足规定的非随机采样条件和温和的不相干假设，所提出的LRTC-TIDT模型实现精确恢复，在各种非随机缺失模式下的仿真实验证实。此外，LRTC-TIDT在多个现实任务中始终优于现有的基于张量的方法，包括网络流量重建，城市交通估计和温度场预测。我们的实现可以在https://github.com/HaoShu2000/LRTC-TIDT上公开获得。
摘要：Non-random missing data is a ubiquitous yet undertreated flaw in multidimensional time series, fundamentally threatening the reliability of data-driven analysis and decision-making. Pure low-rank tensor completion, as a classical data recovery method, falls short in handling non-random missingness, both methodologically and theoretically. Hankel-structured tensor completion models provide a feasible approach for recovering multidimensional time series with non-random missing patterns. However, most Hankel-based multidimensional data recovery methods both suffer from unclear sources of Hankel tensor low-rankness and lack an exact recovery theory for non-random missing data. To address these issues, we propose the temporal isometric delay-embedding transform, which constructs a Hankel tensor whose low-rankness is naturally induced by the smoothness and periodicity of the underlying time series. Leveraging this property, we develop the \textit{Low-Rank Tensor Completion with Temporal Isometric Delay-embedding Transform} (LRTC-TIDT) model, which characterizes the low-rank structure under the \textit{Tensor Singular Value Decomposition} (t-SVD) framework. Once the prescribed non-random sampling conditions and mild incoherence assumptions are satisfied, the proposed LRTC-TIDT model achieves exact recovery, as confirmed by simulation experiments under various non-random missing patterns. Furthermore, LRTC-TIDT consistently outperforms existing tensor-based methods across multiple real-world tasks, including network flow reconstruction, urban traffic estimation, and temperature field prediction. Our implementation is publicly available at https://github.com/HaoShu2000/LRTC-TIDT.
【20】Assessing Neuromorphic Computing for Fingertip Force Decoding from Electromyography
标题：评估来自肌电的指尖力解码的神经形态计算
链接
：https://arxiv.org/abs/2512.10179
备注：5 pages, 6 figures. Poster included as ancillary file (IEEE_NER2025_NeuromorphicEMG_poster.pdf). Presented at IEEE EMBS NER 2025, also at NC State College of Engineering Applied AI Symposium and NC State ECE Graduate Research Symposium (tied for Best Poster)
摘要：高密度表面肌电图（HD-sEMG）为辅助和康复控制提供了一种非侵入性神经接口，但将神经活动映射到用户运动意图仍然具有挑战性。我们评估了尖峰神经网络（SNN）作为神经形态结构对时间卷积网络（TCN）解码指尖力从运动单元（MU）从HD-sEMG发射。数据收集自具有两个前臂电极阵列的单个参与者（10次试验）;通过基于FastICA的分解获得MU活动，并在具有端到端因果卷积的重叠窗口上训练模型。在保留试验中，TCN达到4.44%MVC RMSE（Pearson r = 0.974），而SNN达到8.25%MVC（r = 0.922）。虽然TCN更准确，但我们认为SNN是一个现实的神经形态基线，可以通过适度的架构和超参数改进来缩小这一差距。
摘要：High-density surface electromyography (HD-sEMG) provides a noninvasive neural interface for assistive and rehabilitation control, but mapping neural activity to user motor intent remains challenging. We assess a spiking neural network (SNN) as a neuromorphic architecture against a temporal convolutional network (TCN) for decoding fingertip force from motor-unit (MU) firing derived from HD-sEMG. Data were collected from a single participant (10 trials) with two forearm electrode arrays; MU activity was obtained via FastICA-based decomposition, and models were trained on overlapping windows with end-to-end causal convolutions. On held-out trials, the TCN achieved 4.44% MVC RMSE (Pearson r = 0.974) while the SNN achieved 8.25% MVC (r = 0.922). While the TCN was more accurate, we view the SNN as a realistic neuromorphic baseline that could close much of this gap with modest architectural and hyperparameter refinements.
【21】Semantic-Aware Confidence Calibration for Automated Audio Captioning
标题：自动音频字幕的语义感知置信度校准
链接：https://arxiv.org/abs/2512.10170
备注：5 pages, 2 figures
摘要：自动化音频字幕模型经常产生过度自信的预测，而不管语义准确性，限制了它们在部署中的可靠性。这种缺陷源于两个因素：基于n-gram重叠的评估指标无法捕获语义正确性，以及缺乏校准的置信度估计。我们提出了一个框架，解决了这两个限制，通过集成到音频字幕的置信度预测和重新定义的正确性，通过语义相似性。我们的方法增加了一个基于耳语的音频字幕模型与学习的信心预测头，估计解码器隐藏状态的不确定性。我们采用CLAP音频文本嵌入和句子Transformer相似性（FENSE）来定义语义正确性，从而实现反映真实字幕质量而不是表面级文本重叠的预期校准误差（ECE）计算。Clotho v2上的实验表明，与贪婪解码基线（ECE为0.488）相比，具有语义评估的置信度引导的波束搜索实现了显著改进的校准（基于CLAP的ECE为0.071），同时提高了标准度量的字幕质量。我们的研究结果表明，语义相似性提供了一个更有意义的基础，在音频字幕比传统的n-gram指标的信心校准。
摘要：Automated audio captioning models frequently produce overconfident predictions regardless of semantic accuracy, limiting their reliability in deployment. This deficiency stems from two factors: evaluation metrics based on n-gram overlap that fail to capture semantic correctness, and the absence of calibrated confidence estimation. We present a framework that addresses both limitations by integrating confidence prediction into audio captioning and redefining correctness through semantic similarity. Our approach augments a Whisper-based audio captioning model with a learned confidence prediction head that estimates uncertainty from decoder hidden states. We employ CLAP audio-text embeddings and sentence transformer similarities (FENSE) to define semantic correctness, enabling Expected Calibration Error (ECE) computation that reflects true caption quality rather than surface-level text overlap. Experiments on Clotho v2 demonstrate that confidence-guided beam search with semantic evaluation achieves dramatically improved calibration (CLAP-based ECE of 0.071) compared to greedy decoding baselines (ECE of 0.488), while simultaneously improving caption quality across standard metrics. Our results establish that semantic similarity provides a more meaningful foundation for confidence calibration in audio captioning than traditional n-gram metrics.
【22】Rethinking Causal Discovery Through the Lens of Exchangeability
标题：从可交换性的角度重新思考因果关系的发现
链接：https://arxiv.org/abs/2512.10152
备注：37 pages, 4 figures
摘要：因果发现方法传统上是在两种不同的制度下发展起来的：独立和同分布（i.i.d.）和时间序列数据，每一个都由单独的建模假设支配。在本文中，我们认为，i.i.d.背景可以而且应该根据交换性来重新构建，这是一个严格意义上更普遍的对称原则。我们提出了这种重构的影响，以及两个核心论点：（1）一个概念性的论点，基于扩展的依赖性实验因果推理的因果发现的交换;（2）一个经验性的论点，表明许多现有的i.i.d.数据发现方法是基于交换假设的，并且唯一广泛广泛使用的真实世界“i.i.d.“基准（图宾根数据集）主要由可交换的（而不是i.i.d.）例子.基于这一认识，我们引入了一个新的合成数据集，它只强制执行交换假设，而不强制执行更强的i.i.d.。假设我们表明，我们的可交换的合成数据集反映了现实世界的统计结构“i.i.d.”。“数据集比所有其他i.i.d.合成数据集。此外，我们通过提出一种基于神经网络的cabinet发现算法来证明该数据集的预测能力，该算法专门在我们的合成数据集上进行训练，并且与其他最先进的i.i.d.现实世界的基准。
摘要：Causal discovery methods have traditionally been developed under two distinct regimes: independent and identically distributed (i.i.d.) and timeseries data, each governed by separate modelling assumptions. In this paper, we argue that the i.i.d. setting can and should be reframed in terms of exchangeability, a strictly more general symmetry principle. We present the implications of this reframing, alongside two core arguments: (1) a conceptual argument, based on extending the dependency of experimental causal inference on exchangeability to causal discovery; and (2) an empirical argument, showing that many existing i.i.d. causal-discovery methods are predicated on exchangeability assumptions, and that the sole extensive widely-used real-world "i.i.d." benchmark (the Tübingen dataset) consists mainly of exchangeable (and not i.i.d.) examples. Building on this insight, we introduce a novel synthetic dataset that enforces only the exchangeability assumption, without imposing the stronger i.i.d. assumption. We show that our exchangeable synthetic dataset mirrors the statistical structure of the real-world "i.i.d." dataset more closely than all other i.i.d. synthetic datasets. Furthermore, we demonstrate the predictive capability of this dataset by proposing a neural-network-based causal-discovery algorithm trained exclusively on our synthetic dataset, and which performs similarly to other state-of-the-art i.i.d. methods on the real-world benchmark.
【23】Push Smarter, Not Harder: Hierarchical RL-Diffusion Policy for Efficient Nonprehensile Manipulation
标题：更聪明地推动，而不是更困难：高效非可旋转操纵的分层RL扩散政策
链接：https://arxiv.org/abs/2512.10099
备注：8 pages, 8 figures
摘要：由于复杂的接触动力学和长期规划要求，非接触操纵，如在杂乱的环境中推动物体，提出了一个具有挑战性的控制问题。在这项工作中，我们提出了HeRD，一个层次化的强化学习扩散策略，将推任务分解为两个层次：高级目标选择和低级轨迹生成。我们采用高级强化学习（RL）代理来选择中间空间目标，并采用低级目标条件扩散模型来生成可行的，有效的轨迹来达到它们。 这种架构将RL的长期奖励最大化行为与扩散模型的生成能力相结合。我们在2D模拟环境中评估了我们的方法，并表明它在成功率、路径效率和跨多种环境配置的泛化方面优于最先进的基线。我们的研究结果表明，分层控制与生成的低层次规划是一个有前途的方向，可扩展的，目标导向的nonprojectile操纵。代码、文档和训练模型可在https://github.com/carosteven/HeRD上找到。
摘要 ：Nonprehensile manipulation, such as pushing objects across cluttered environments, presents a challenging control problem due to complex contact dynamics and long-horizon planning requirements. In this work, we propose HeRD, a hierarchical reinforcement learning-diffusion policy that decomposes pushing tasks into two levels: high-level goal selection and low-level trajectory generation. We employ a high-level reinforcement learning (RL) agent to select intermediate spatial goals, and a low-level goal-conditioned diffusion model to generate feasible, efficient trajectories to reach them. This architecture combines the long-term reward maximizing behaviour of RL with the generative capabilities of diffusion models. We evaluate our method in a 2D simulation environment and show that it outperforms the state-of-the-art baseline in success rate, path efficiency, and generalization across multiple environment configurations. Our results suggest that hierarchical control with generative low-level planning is a promising direction for scalable, goal-directed nonprehensile manipulation. Code, documentation, and trained models are available: https://github.com/carosteven/HeRD.
【24】Fuzzy Hierarchical Multiplex
标题：模糊分层多重
链接：https://arxiv.org/abs/2512.09976
备注：11 pages, 2 figures, 1 double figure, 1 table, 12 references. This will be part of my PhD dissertation and it is a White paper-theoretical framewor. As is, it s meant for a basis that will be later used to further developed an FHM. It might not be math-logic related and I am willing to change it, I just felt that it belonged to mathematical modeling. Yours truly, AK
摘要：提出了一种扩展FCM因果关系的模糊优化框架。该模型利用动态将数据映射到度量，并创建一个框架，使用多路复用来检查概念的逻辑含义和层次结构。在此基础上，阐述了该框架的主要目标和定位，并举例说明了该框架在服务流程设计中的信息传递服务优化。最后，一个彻底的分析FHM包括这是做以下的逻辑步骤，在一个简单而优雅的方式。
摘要：A new fuzzy optimization framework that extends FCM causality is proposed. This model utilizes the dynamics to map data into metrics and create a framework that examines logical implication and hierarchy of concepts using a multiplex. Moreover, this is a white-theoretical paper introducing the framework and analyzing the logic and math behind it. Upon this extension the main objectives and the orientation of this framework is expounded and exemplified; this framework is meant for service optimization of information transmission in service process design. Lastly, a thorough analysis of the FHM is included which is done following the logical steps in a simple and elegant manner.
【25】TDC-Cache: A Trustworthy Decentralized Cooperative Caching Framework for Web3.0
标题：TDC-Cache：一个面向Web3.0的可信分布式协同缓存框架
链接：https://arxiv.org/abs/2512.09961
摘要：Web3.0的快速发展正在将互联网从中心化结构转变为去中心化结构，这使用户对自己的数据拥有前所未有的自主权。然而，在Web3.0中分散式数据访问的背景下，必须应对冗余数据复制引起的效率问题，以及数据不一致引起的安全漏洞。为了应对这些挑战，我们开发了一个可信的分散式合作缓存（TDC-Cache）框架的Web3.0，以确保高效的缓存和增强系统的弹性对抗性威胁。该框架具有两层架构，其中分散式Oracle网络（DON）层作为分散式缓存的可信中介平台，桥接来自分散式存储的内容和来自用户的内容请求。针对Web3.0网络拓扑结构和数据流的复杂性，提出了一种基于深度强化学习的分布式缓存（DRL-DC）的TDC-Cache，用于动态优化分布式Oracle的缓存策略。此外，我们开发了一个证明合作学习（PoCL）的共识，以保持DON内的分散缓存决策的一致性。实验结果表明，与现有方法相比，该框架将平均访问延迟降低了20%，缓存命中率最多提高了18%，平均成功共识率提高了10%。总的来说，本文是对Web3.0的分散式缓存框架和策略的第一次尝试。
摘要：The rapid growth of Web3.0 is transforming the Internet from a centralized structure to decentralized, which empowers users with unprecedented self-sovereignty over their own data. However, in the context of decentralized data access within Web3.0, it is imperative to cope with efficiency concerns caused by the replication of redundant data, as well as security vulnerabilities caused by data inconsistency. To address these challenges, we develop a Trustworthy Decentralized Cooperative Caching (TDC-Cache) framework for Web3.0 to ensure efficient caching and enhance system resilience against adversarial threats. This framework features a two-layer architecture, wherein the Decentralized Oracle Network (DON) layer serves as a trusted intermediary platform for decentralized caching, bridging the contents from decentralized storage and the content requests from users. In light of the complexity of Web3.0 network topologies and data flows, we propose a Deep Reinforcement Learning-Based Decentralized Caching (DRL-DC) for TDC-Cache to dynamically optimize caching strategies of distributed oracles. Furthermore, we develop a Proof of Cooperative Learning (PoCL) consensus to maintain the consistency of decentralized caching decisions within DON. Experimental results show that, compared with existing approaches, the proposed framework reduces average access latency by 20%, increases the cache hit rate by at most 18%, and improves the average success consensus rate by 10%. Overall, this paper serves as a first foray into the investigation of decentralized caching framework and strategy for Web3.0.
【26】ZK-APEX: Zero-Knowledge Approximate Personalized Unlearning with Executable Proofs
标题：ZK-APEX：具有可执行证明的零知识近似个性化取消学习
链接：https://arxiv.org/abs/2512.09953
摘要：机器非学习旨在从训练模型中消除特定数据点的影响，以满足隐私，版权和安全要求。在实际部署中，提供商将全局模型分发到许多边缘设备，其中每个客户端使用私有数据对模型进行个性化。当发出删除请求时，客户端可能会忽略它或错误地声称合规，并且提供商无法检查其参数或数据。这使得验证变得困难，特别是因为个性化模型必须忘记目标样本，同时保留本地效用，并且验证必须在边缘设备上保持轻量级。 我们介绍ZK APEX，一个zero-shot个性化的unlearning方法，直接在个性化的模型上操作，无需再训练。ZK APEX将提供方的稀疏掩码与客户端的小组OBS补偿步骤相结合，使用分块经验Fisher矩阵来创建为低开销而设计的曲率感知更新。与Halo2零知识证明相结合，它使提供商能够验证是否应用了正确的非学习转换，而不会泄露任何私人数据或个性化参数。 在Vision Transformer分类任务中，ZK APEX几乎可以恢复所有的个性化准确性，同时有效地删除目标信息。应用于在代码数据上训练的OPT125M生成模型，它可以恢复约70%的原始准确度。ViT案例的证明生成大约在两个小时内完成，比基于再训练的检查快一千万倍以上，内存使用量不到1GB，证明大小约为400兆字节。这些结果显示了第一个在边缘设备上可验证的个性化非学习的实用框架。
摘要 ：Machine unlearning aims to remove the influence of specific data points from a trained model to satisfy privacy, copyright, and safety requirements. In real deployments, providers distribute a global model to many edge devices, where each client personalizes the model using private data. When a deletion request is issued, clients may ignore it or falsely claim compliance, and providers cannot check their parameters or data. This makes verification difficult, especially because personalized models must forget the targeted samples while preserving local utility, and verification must remain lightweight on edge devices. We introduce ZK APEX, a zero-shot personalized unlearning method that operates directly on the personalized model without retraining. ZK APEX combines sparse masking on the provider side with a small Group OBS compensation step on the client side, using a blockwise empirical Fisher matrix to create a curvature-aware update designed for low overhead. Paired with Halo2 zero-knowledge proofs, it enables the provider to verify that the correct unlearning transformation was applied without revealing any private data or personalized parameters. On Vision Transformer classification tasks, ZK APEX recovers nearly all personalization accuracy while effectively removing the targeted information. Applied to the OPT125M generative model trained on code data, it recovers around seventy percent of the original accuracy. Proof generation for the ViT case completes in about two hours, more than ten million times faster than retraining-based checks, with less than one gigabyte of memory use and proof sizes around four hundred megabytes. These results show the first practical framework for verifiable personalized unlearning on edge devices.
【27】Echo-CoPilot: A Multi-View, Multi-Task Agent for Echocardiography Interpretation and Reporting
标题：Echo-CoPilot：一种用于超声心动图解释和报告的多视图、多任务代理
链接：https://arxiv.org/abs/2512.09944
摘要：超声心动图是当代心血管护理的核心，但完整的研究解释仍然是一个认知要求很高的多视图任务，仍然是手动执行。虽然最近的超声心动图基础模型可以在视图分类、分割或疾病预测等单个感知子任务上实现强大的性能，但它们通常孤立地操作，并且不提供统一的临床一致性评估。在这项工作中，我们介绍了Echo-CoPilot，一个多视图，多任务代理，使用一个大型的语言模型来编排一套专门的超声心动图工具。在ReAct式循环中，代理分解临床医生查询，调用视图识别，心脏结构分割，测量和疾病预测以及报告合成的工具，并将其输出集成到指南感知答案和叙述性摘要中。我们在公共MIMIC-EchoQA基准测试中评估了Echo-CoPilot，其准确率为50.8%，优于通用和生物医学视频视觉语言模型。定性分析进一步表明，该代理商利用定量测量和生理背景来解决临床决策阈值附近的挑战性病例，例如边缘左心室肥大或心包积液严重程度。该代码将在论文被接受后发布。
摘要：Echocardiography is central to contemporary cardiovascular care, but full-study interpretation remains a cognitively demanding, multi-view task that is still performed manually. While recent foundation models for echocardiography can achieve strong performance on individual perceptual subtasks such as view classification, segmentation, or disease prediction, they typically operate in isolation and do not provide a unified, clinically coherent assessment. In this work, we introduce Echo-CoPilot, a multi-view, multi-task agent that uses a large language model to orchestrate a suite of specialized echocardiography tools. Within a ReAct-style loop, the agent decomposes clinician queries, invokes tools for view recognition, cardiac structure segmentation, measurement and disease prediction, and report synthesis, and integrates their outputs into guideline-aware answers and narrative summaries. We evaluate Echo-CoPilot on the public MIMIC-EchoQA benchmark, where it achieves an accuracy of 50.8\%, outperforming both general-purpose and biomedical video vision-language models. Qualitative analyses further show that the agent leverages quantitative measurements and physiologic context to resolve challenging cases near clinical decision thresholds, such as borderline left ventricular hypertrophy or pericardial effusion severity. The code will be released upon acceptance of the paper.
【28】Norm-Governed Multi-Agent Decision-Making in Simulator-Coupled Environments:The Reinsurance Constrained Multi-Agent Simulation Process (R-CMASP)
标题：模拟器耦合环境中的常态管理多智能体决策：再保险约束多智能体模拟过程（R-CMASP）
链接：https://arxiv.org/abs/2512.09939
摘要：再保险决策具有激励多代理模型的核心结构属性：分布式和非对称信息，部分可观察性，异构认知责任，模拟器驱动的环境动态，以及有约束力的审慎和监管约束。确定性的工作流自动化不能满足这些要求，因为它缺乏认识的灵活性，合作协调机制，和规范敏感的行为所需的机构风险转移。 我们提出了再保险约束多主体模拟过程（R-CMASP），一个正式的模型，扩展了随机博弈和Dec-POMDPs，增加了三个缺失的元素：（i）模拟器耦合的过渡动态接地灾难，资本和投资组合引擎;（ii）角色专用代理结构化的可观察性，信念更新，和类型的通信;和（iii）一个规范的可行性层，将偿付能力、监管和组织规则编码为对联合行动的可接受性约束。 使用基于LLM的代理与工具访问和类型化消息协议，我们在一个域校准的合成环境中显示，与确定性自动化或单片LLM基线相比，受管理的多代理协调产生更稳定，连贯和规范遵守的行为-减少定价差异，提高资本效率，并增加子句解释的准确性。嵌入审慎规范作为容许性约束，并将通信结构化为类型化行为，可显著增强均衡稳定性。 总的来说，结果表明，规范，模拟器驱动的决策环境是最自然的建模为规范管理，模拟器耦合的多智能体系统。
摘要：Reinsurance decision-making exhibits the core structural properties that motivate multi-agent models: distributed and asymmetric information, partial observability, heterogeneous epistemic responsibilities, simulator-driven environment dynamics, and binding prudential and regulatory constraints. Deterministic workflow automation cannot meet these requirements, as it lacks the epistemic flexibility, cooperative coordination mechanisms, and norm-sensitive behaviour required for institutional risk-transfer. We propose the Reinsurance Constrained Multi-Agent Simulation Process (R-CMASP), a formal model that extends stochastic games and Dec-POMDPs by adding three missing elements: (i) simulator-coupled transition dynamics grounded in catastrophe, capital, and portfolio engines; (ii) role-specialized agents with structured observability, belief updates, and typed communication; and (iii) a normative feasibility layer encoding solvency, regulatory, and organizational rules as admissibility constraints on joint actions. Using LLM-based agents with tool access and typed message protocols, we show in a domain-calibrated synthetic environment that governed multi-agent coordination yields more stable, coherent, and norm-adherent behaviour than deterministic automation or monolithic LLM baselines--reducing pricing variance, improving capital efficiency, and increasing clause-interpretation accuracy. Embedding prudential norms as admissibility constraints and structuring communication into typed acts measurably enhances equilibrium stability. Overall, the results suggest that regulated, simulator-driven decision environments are most naturally modelled as norm-governed, simulator-coupled multi-agent systems.
【29】Hermitian Yang--Mills connections on general vector bundles: geometry and physical Yukawa couplings
标题：埃尔米特·杨--一般载体束上的米尔斯联系：几何和物理汤川耦合
链接：https://arxiv.org/abs/2512.10907
备注：51 pages, Associated code open--sourced at https://github.com/Justin-Tan/cymyc
摘要：我们计算全纯向量丛$V$通过一个交替的优化过程建立在几何机器学习的埃尔米特杨米尔斯方程的解决方案。所提出的方法是完全一般的秩和结构组的$V$，只需要能够枚举一个基础的全球部分为一个给定的捆绑。这使我们能够计算在一个广泛的类杂合弦紧化的物理规范化汤川耦合。利用这种方法，我们对包含非阿贝尔结构群规范丛的杂合紧化进行了完整的计算。
摘要：We compute solutions to the Hermitian Yang-Mills equations on holomorphic vector bundles $V$ via an alternating optimisation procedure founded on geometric machine learning. The proposed method is fully general with respect to the rank and structure group of $V$, requiring only the ability to enumerate a basis of global sections for a given bundle. This enables us to compute the physically normalised Yukawa couplings in a broad class of heterotic string compactifications. Using this method, we carry out this computation in full for a heterotic compactification incorporating a gauge bundle with non-Abelian structure group.
【30】Quantum Approaches to Urban Logistics: From Core QAOA to Clustered Scalability
标题：城市物流的量子方法：从核心QAOA到可扩展性
链接：https://arxiv.org/abs/2512.10813
摘要 ：旅行商问题（TSP）是组合优化中的一个基本问题，广泛应用于物流和运输领域。随着TSP实例规模的增长，传统算法往往难以在合理的时间范围内产生高质量的解决方案。本研究探讨量子近似最佳化演算法（QAOA），一种混合量子-古典方法，在现实限制下解决TSP的潜力。我们采用了基于QUBO的TSP公式，该公式集成了反映运营条件的现实物流约束，如车辆容量、道路可达性和时间窗口，同时确保与当前量子硬件的限制兼容。我们的实验是在模拟环境中使用高性能计算（HPC）资源来评估QAOA在不同问题大小和量子电路深度上的性能。为了提高可扩展性，我们提出了聚类QAOA（Cl-QAOA），一种结合经典机器学习和QAOA的混合方法。这种方法将大型TSP实例分解为较小的子问题，即使在量子位数有限的设备上也可以进行量子优化。结果提供了一个全面的评价QAOA的优势和局限性，在解决约束TSP的情况下。这项研究推进了量子优化，并为未来的大规模应用奠定了基础。
摘要：The Traveling Salesman Problem (TSP) is a fundamental challenge in combinatorial optimization, widely applied in logistics and transportation. As the size of TSP instances grows, traditional algorithms often struggle to produce high-quality solutions within reasonable timeframes. This study investigates the potential of the Quantum Approximate Optimization Algorithm (QAOA), a hybrid quantum-classical method, to solve TSP under realistic constraints. We adopt a QUBO-based formulation of TSP that integrates real-world logistical constraints reflecting operational conditions, such as vehicle capacity, road accessibility, and time windows, while ensuring compatibility with the limitations of current quantum hardware. Our experiments are conducted in a simulated environment using high-performance computing (HPC) resources to assess QAOA's performance across different problem sizes and quantum circuit depths. In order to improve scalability, we propose clustering QAOA (Cl-QAOA), a hybrid approach combining classical machine learning with QAOA. This method decomposes large TSP instances into smaller sub-problems, making quantum optimization feasible even on devices with a limited number of qubits. The results offer a comprehensive evaluation of QAOA's strengths and limitations in solving constrained TSP scenarios. This study advances quantum optimization and lays groundwork for future large-scale applications.
【31】PMB-NN: Physiology-Centred Hybrid AI for Personalized Hemodynamic Monitoring from Photoplethysmography
标题：PMB-NN：以生理学为中心的混合AI，用于通过光电体积图进行个性化血流动力学监测
链接：https://arxiv.org/abs/2512.10745
摘要：连续监测血压（BP）和血流动力学参数（如外周阻力（R）和动脉顺应性（C））对于早期血管功能障碍检测至关重要。虽然光电体积描记（PPG）可穿戴设备已经普及，但现有的BP估计数据驱动方法缺乏可解释性。我们在血压估计中推进了我们之前提出的以生理学为中心的混合AI方法-基于生理模型的神经网络（PMB-NN），该方法将深度学习与由R和C作为物理约束参数化的基于2元素Windkessel的模型相结合。PMB-NN模型使用PPG衍生的定时特征以受试者特定的方式进行训练，而人口统计学信息用于推断中间变量：心输出量。我们在10名健康成年人身上验证了模型的日常鲁棒性，这些成年人在两天内进行静态和骑自行车活动，以深度学习（DL）模型（FCNN，CNN-LSTM，Transformer）和基于Windkessel的独立生理模型（PM）为基准。验证从三个方面进行：准确性、可解释性和可解释性。PMB-NN实现了与DL基准相当的收缩压准确度（MAE：7.2 mmHg），舒张压性能（MAE：3.9 mmHg）低于DL模型。然而，PMB-NN表现出更高的生理兼容性比DL基线和PM，这表明混合架构统一和增强生理原理和数据驱动技术各自的优点。除了BP，PMB-NN在训练过程中识别了R（ME：0.15 mmHg$\cdot$s/ml）和C（ME：-0.35 ml/mmHg），其准确性与PM相似，表明嵌入的生理约束赋予了混合AI框架的可解释性。这些结果将PMB-NN定位为用于日常血流动力学监测的纯数据驱动方法的平衡的、生理学基础的替代方案。
摘要：Continuous monitoring of blood pressure (BP) and hemodynamic parameters such as peripheral resistance (R) and arterial compliance (C) are critical for early vascular dysfunction detection. While photoplethysmography (PPG) wearables has gained popularity, existing data-driven methods for BP estimation lack interpretability. We advanced our previously proposed physiology-centered hybrid AI method-Physiological Model-Based Neural Network (PMB-NN)-in blood pressure estimation, that unifies deep learning with a 2-element Windkessel based model parameterized by R and C acting as physics constraints. The PMB-NN model was trained in a subject-specific manner using PPG-derived timing features, while demographic information was used to infer an intermediate variable: cardiac output. We validated the model on 10 healthy adults performing static and cycling activities across two days for model's day-to-day robustness, benchmarked against deep learning (DL) models (FCNN, CNN-LSTM, Transformer) and standalone Windkessel based physiological model (PM). Validation was conducted on three perspectives: accuracy, interpretability and plausibility. PMB-NN achieved systolic BP accuracy (MAE: 7.2 mmHg) comparable to DL benchmarks, diastolic performance (MAE: 3.9 mmHg) lower than DL models. However, PMB-NN exhibited higher physiological plausibility than both DL baselines and PM, suggesting that the hybrid architecture unifies and enhances the respective merits of physiological principles and data-driven techniques. Beyond BP, PMB-NN identified R (ME: 0.15 mmHg$\cdot$s/ml) and C (ME: -0.35 ml/mmHg) during training with accuracy similar to PM, demonstrating that the embedded physiological constraints confer interpretability to the hybrid AI framework. These results position PMB-NN as a balanced, physiologically grounded alternative to purely data-driven approaches for daily hemodynamic monitoring.
【32】Maximum Risk Minimization with Random Forests
标题：使用随机森林最大限度地降低风险
链接：https://arxiv.org/abs/2512.10445
备注：47 pages, 13 figures
摘要：我们考虑一个回归设置，在不同的数据分布建模的不同环境中收集观察结果。分布外（OOD）泛化领域旨在设计更好地泛化到测试环境的方法，这些测试环境的分布与训练期间观察到的分布不同。其中一种方法是将环境中的最大风险最小化，我们称之为MaxRM（Maximum Risk Minimization）。在这项工作中，我们介绍了基于MaxRM的原则的随机森林的变种。我们提供计算效率的算法，并证明我们的主要方法的统计一致性。我们提出的方法可以用于以下三种风险中的每一种：均方误差，负回报（与解释方差相关）和后悔（量化相对于最佳预测的超额风险）。对于MaxRM与遗憾的风险，我们证明了一个新的样本外保证看不见的测试分布。最后，我们评估所提出的方法模拟和真实世界的数据。
摘要：We consider a regression setting where observations are collected in different environments modeled by different data distributions. The field of out-of-distribution (OOD) generalization aims to design methods that generalize better to test environments whose distributions differ from those observed during training. One line of such works has proposed to minimize the maximum risk across environments, a principle that we refer to as MaxRM (Maximum Risk Minimization). In this work, we introduce variants of random forests based on the principle of MaxRM. We provide computationally efficient algorithms and prove statistical consistency for our primary method. Our proposed method can be used with each of the following three risks: the mean squared error, the negative reward (which relates to the explained variance), and the regret (which quantifies the excess risk relative to the best predictor). For MaxRM with regret as the risk, we prove a novel out-of-sample guarantee over unseen test distributions. Finally, we evaluate the proposed methods on both simulated and real-world data.
【33】Diffusion differentiable resampling
标题：扩散可微重采样
链接：https://arxiv.org/abs/2512.10401
摘要：本文关注的是序列蒙特卡罗（例如，粒子滤波）。我们提出了一种新的信息检索方法，即时路径可微，基于合奏得分扩散模型。我们证明了我们的扩散恢复方法提供了一个一致的估计恢复分布，我们通过实验表明，它优于国家的最先进的微分恢复方法时，用于随机滤波和参数估计。
摘要：This paper is concerned with differentiable resampling in the context of sequential Monte Carlo (e.g., particle filtering). We propose a new informative resampling method that is instantly pathwise differentiable, based on an ensemble score diffusion model. We prove that our diffusion resampling method provides a consistent estimate to the resampling distribution, and we show by experiments that it outperforms the state-of-the-art differentiable resampling methods when used for stochastic filtering and parameter estimation.
【34】Residual subspace evolution strategies for nonlinear inverse problems
标题：非线性反问题的残差子空间演化策略
链接：https://arxiv.org/abs/2512.10325
摘要：非线性逆问题通常具有噪音、不可微或昂贵的残差评估，这使得基于雅可比的求解器不可靠。流行的无导数优化器，如自然进化策略（NES）或鲍威尔的NEWUOA仍然假设平滑或花费许多评估来保持稳定性。包络卡尔曼反演（EKI）依赖于需要预处理的经验协方差，并且与残差维数的比例关系很差。 我们引入残差子空间演化策略（RSES），一个无导数求解器，该求解器对当前的高斯探测器进行采样，从它们的差异中构建一个仅残差的代理，并通过最小二乘求解重新组合探测器，从而在不形成雅可比矩阵或协方差的情况下获得最佳更新。每次迭代的代价是$k+1$$余值计算，其中$k \ll n$对于$n$维问题，线性代数开销为$O（k^3）$。 校准，回归和反卷积问题的基准表现出一致的失配减少确定性和随机设置。RSES匹配或超过xNES和NEWUOA，同时在匹配的评估预算下与EKI保持竞争力，特别是当平滑或协方差假设失败时。
摘要：Nonlinear inverse problems often feature noisy, non-differentiable, or expensive residual evaluations that make Jacobian-based solvers unreliable. Popular derivative-free optimizers such as natural evolution strategies (NES) or Powell's NEWUOA still assume smoothness or expend many evaluations to maintain stability. Ensemble Kalman inversion (EKI) relies on empirical covariances that require preconditioning and scale poorly with residual dimension. We introduce residual subspace evolution strategies (RSES), a derivative-free solver that samples Gaussian probes around the current iterate, builds a residual-only surrogate from their differences, and recombines the probes through a least-squares solve yielding an optimal update without forming Jacobians or covariances. Each iteration costs $k+1$ residual evaluations, where $k \ll n$ for $n$-dimensional problems, with $O(k^3)$ linear algebra overhead. Benchmarks on calibration, regression, and deconvolution problems demonstrate consistent misfit reduction in both deterministic and stochastic settings. RSES matches or surpasses xNES and NEWUOA while staying competitive with EKI under matched evaluation budgets, particularly when smoothness or covariance assumptions fail.
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/190413
本文地址：http://www.python88.com/topic/190413