机器学习学术速递[7.21]

AI 总结：研究针对FTaaS平台削弱模型安全对齐的问题，提出TRACE框架，通过模拟有害轨迹生成损坏状态，优化插件补丁，在保持效用的同时恢复模型安全，在多基准测试和模型上占据安全 - 效用前沿。

链接：https://arxiv.org/abs/2607.16242

机构：LERo

作者：Changyue Li, Jiaming He, Youliang Yuan, Jialin Wu, Boxi Yu, Zhicong Huang, Pinjia He

英文摘要：Fine-Tuning-as-a-Service (FTaaS) platforms let users train large language models (LLMs) on customized tasks, but this pipeline could erode models' safety alignment. In practice, service providers need to recover models' safety without re-running full alignment, or destroying the utility gained from customized tasks. A line of existing work refers to model parameter merging, which adds a safety patch on the fine-tuned model parameters to shift the model away from unsafe tendencies. However, this merging-based paradigm is fundamentally bottlenecked by task-safety update entanglement: downstream task updates and the safety patch often overlap in their dominant directions, so the merge strength is intrinsically hard to calibrate. If the safety vector is scaled too weakly, harmful components could still dominate, preventing the model from returning to a safe region; if it is scaled too aggressively, it suppresses task-relevant directions and degrades utility. To solve this problem, we shift the focus of merging-based methods from designing online merging operators to offline patch learning, and seek a safety patch that minimally interferes with task-relevant directions while retaining decisive control over unsafe behaviors. We propose TRACE, a trajectory-based safety patch learning framework that (i) simulates harmful tuning trajectories to generate progressively corrupted states, and (ii) optimizes a plug-in patch to recover safety while maintaining utility across varying corrupted base states. Across six benchmarks and two models, TRACE consistently dominates the safety-utility frontier. TRACE reaches nearly 100% safety on all settings, while maintaining comparable utility to the undefended fine-tuned model.

AI 总结：研究针对多模态生存分析中传统方法的局限，提出AdaSurvMamba框架，含DSIR模块动态调制跨模态交互强度、SAS模块重组令牌成连续序列，实验证明该框架在五个TCGA队列上比现有方法有优势。

链接：https://arxiv.org/abs/2607.16260

机构：Dalian University of Technology（大连理工大学）； University of Alberta（阿尔伯塔大学）； Yale University（耶鲁大学）

作者：Jialong Zhong, Tingwei Liu, Baokun Yue, Jingjing Li, Yongri Piao, Miao Zhang, Leiye Liu, Jiahong Jiang, Wei Ji, Huchuan Lu

英文摘要：Multimodal survival analysis utilizing whole slide images (WSIs) and genomic profiles is fundamental for cancer prognosis. Recently, state-space models like Mamba have emerged as powerful tools for sequence modeling. However, translating this success to complex multimodal tasks is hindered by two critical limitations. First, conventional fusion strategies assume a static multimodal interaction strength, ignoring the fluctuating diagnostic importance of each modality across different patients and local regions. Second, the standard Mamba architecture processes tokens along predefined physical paths. This rigid scanning disrupts the semantic continuity of spatially scattered medical features and exacerbates long-range decay. To address these challenges, we introduce AdaSurvMamba as a novel adaptive framework for multimodal survival analysis. The framework features a Dual-Scale Importance-Aware Reconstruction (DSIR) module to dynamically modulate cross-modal interaction strength. It evaluates diagnostic importance at both the sequence and token levels to reconstruct the input representations. Furthermore, we propose a Semantic Aggregation Scanning (SAS) module to overcome contextual fragmentation. The SAS module dynamically reorganizes discrete tokens into semantically continuous sequences via a shared prototype pool. It explicitly modulates the state transition step size using global modality context and semantic priors to adaptively control the information absorption rate. Experiments across five TCGA cohorts demonstrate consistent gains over existing methods. Code is available at this https URL.

AI 总结：针对高级深度学习库抽象神经网络内部机制的问题，本文从零实现自包含神经网络框架，涵盖关键组件。该框架兼具教学功能，应用于多类分类任务性能强大，其设计和模块化使其成为教育及研究的可靠基线。

链接：https://arxiv.org/abs/2607.16682

机构：University of Sydney（悉尼大学）

作者：Yuanzhe Jia

英文摘要： The widespread adoption of high-level deep learning libraries, while accelerating model development, has increasingly abstracted away the internal mechanics of neural networks, creating a gap between practical usage and fundamental understanding. To address this, the paper presents a self-contained neural network framework implemented entirely from scratch -- without relying on automatic differentiation or pre-built deep learning modules. The implementation encompasses all essential components, including multi-layer architectures, diverse activation functions, regularization techniques, and state-of-the-art optimizers. Beyond serving as a pedagogical instrument that demystifies forward/backward propagation, gradient dynamics, and optimization landscapes, the framework demonstrates robust performance when applied to a multi-class classification task, successfully validating its correctness, numerical stability, and generalization across varied configurations. The extensible design and clean modularity further position it as a reliable baseline for educational purposes and future research exploration.

AI 总结：研究针对编码任务的专家混合语言模型，通过五种策略对两个模型进行剪枝，探究能移除多少及哪些专家。发现移除一半专家在代码基准测试无损失，损害多在编码外，但获胜策略因模型而异，还揭示了困惑度等问题及剪枝与量化的关系，强调需 per - model 验证。

链接：https://arxiv.org/abs/2607.16721

作者：Anik Jha

英文摘要：The strongest open-weight coding models are mixture-of-experts (MoE) networks: most of their size comes from large pools of "expert" subnetworks, of which only a few act on any token. That pool is why these models do not fit on the machines most developers own, yet for a user who only wants coding help, most experts encode abilities that will never be invoked. We ask how many experts can be removed, and which, by pruning two recent open-weight MoE models from different families (Qwen3.6-35B-A3B and Gemma-4-26B-A4B) under five selection strategies, judged the way a user would: by whether the model still writes correct code. Half the experts can be removed from either model with no statistically detectable loss on the primary code benchmark, and the damage lands almost entirely on abilities outside coding, the intended trade. But the winning strategy flips between the two models, so a recipe validated on one family cannot be assumed to work on another. We further show that perplexity, the metric much of the pruning literature leans on, can rate a broken model above an intact one; that a lightweight fine-tune recovers about half of what aggressive pruning loses; and that against quantizing the full model to the same memory, pruning wins only where quantization would have to drop below 3 bits per weight. Five attempts to overturn that crossover, with failure criteria fixed in advance (better calibration, guarded selection, causal expert importance, failure attribution, and an agentic evaluation letting each model repair its failures from execution feedback), all leave it standing; the last shows single-shot benchmarks overstate compression penalties broadly, as one repair turn erases the 2-bit quantization penalty entirely. Expert pruning works, but it demands per-model validation on the task the model will actually serve.

AI 总结：研究过渡金属催化反应产率预测难题，提出ChemFusion多模态交叉注意力网络，融合电子特征与3D原子坐标，用交叉注意力机制，在交叉偶联库基准测试中性能出色，还能自主学习识别和惩罚空间位阻，提供物理可解释性。

链接：https://arxiv.org/abs/2607.17033

机构：Duke University（杜克大学）； Georgia Institute of Technology（佐治亚理工学院）

作者：Qiwei Han, Chi Zhou

英文摘要：Forecasting the outcomes of transition-metal-catalyzed reactions is notoriously complex due to the interplay of diverse physical and chemical variables. A persistent computational bottleneck has been effectively merging broad electronic descriptors with the localized, three-dimensional geometry of the reactive site. To bridge this representation gap, we present ChemFusion, a hybrid neural network that fuses conventional electronic features with explicit 3D atomic coordinates. Using a cross-attention mechanism, the model enables global electronic states to dynamically attend to specific spatial constraints within un-pooled molecular point clouds. When benchmarked against a diverse library of cross-couplings, this approach delivers exceptional predictive performance, decisively surpassing traditional single-modality frameworks. Importantly, extracting the attention matrices reveals that the architecture autonomously learns to identify and penalize restrictive steric hindrances. This provides a physically grounded interpretability, demonstrating that spatially aware networks can navigate complex reaction sterics that standard statistical models typically miss.

AI 总结：研究大语言模型训练后处理问题，提出蒸馏强化学习方法，集成教师监督到RL目标，含反向重要性采样等三个组件，能有效转移知识，在家族内和跨家族蒸馏实验中性能大幅优于标准RL和OPD。

链接：https://arxiv.org/abs/2607.17247

机构：College of Elite Engineers, Nankai University（南开大学精英工程师学院）； Zhongguancun Academy（中关村学院）； Beijing Institute of Technology（北京理工大学）； Zhejiang University（浙江大学）； Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）； Harbin Institute of Technology（哈尔滨工业大学）； College of Software, Nankai University（南开大学软件学院）

作者：Chen Wang, Zhaochun Li, Jionghao Bai, Yining Zhang, Hexuan Deng, Ge Lan, Yue Wang

英文摘要：Large language model (LLM) post-training is essential for improving reasoning, adaptation, and alignment. Existing methods mainly follow two paradigms: reinforcement learning (RL) and on-policy distillation (OPD). However, RL relies on coarse-grained outcome supervision, resulting in difficult credit assignment and limited capability to acquire new knowledge. OPD, meanwhile, unconditionally matches teacher logits through KL divergence, which creates a dilemma: similar teachers provide little new knowledge, while substantially different teachers often yield ineffective guidance, largely restricting OPD to within-family distillation. We propose Distilled Reinforcement Learning (Distilled RL), which integrates teacher supervision into the RL objective to provide fine-grained guidance, selectively transfer new knowledge and avoid unconditional imitation. Distilled RL contains three components: reverse importance sampling with clipping, negative sample reset, and sequence-level geometric normalization. Through a concise and interpretable case study, we demonstrate that Distilled RL can effectively transfer previously unavailable knowledge from a teacher model to a student model. Extensive experiments across both within-family and cross-family distillation settings show that Distilled RL substantially outperforms standard RL and OPD in terms of both pass@1 and pass@k. Our code is available at this https URL.

AI 总结：研究“无审查”开放权重模型中删除模型拒绝方向的操作影响，以21600个不确定性决策为探针，对比两个专家混合模型家族的基础与删除模型，发现有多种效应，还发现污染渠道，表明部署此类模型得到的是不同决策者。

链接：https://arxiv.org/abs/2607.17427

机构：huihui-ai（慧慧人工智能）

作者：Aleksander Fafuła

英文摘要：Abliteration - deleting a model's refusal direction from its weights - is the standard recipe behind popular "uncensored" open-weight models. We show the surgery is not clean. As a disposition probe we use 21,600 decisions under uncertainty - weekly up/down calls on 60 Warsaw Stock Exchange equities over 18 weeks, replayed through a frozen pipeline so the decision-layer model is the only variable. The task elicits no refusals at all, so any between-arm delta is pure side effect. Holding provenance constant (official BF16 checkpoints, a single abliteration author, an identical serving stack, one byte-identical frozen prompt), we compare base and abliterated arms of two Mixture-of-Experts families, Gemma-4-26B-A4B-it and Qwen3-30B-A3B-Instruct-2507. Three effects replicate across both families (weeks-clustered bootstrap CIs excluding zero): abliterated models are systematically more optimistic (+12.2 pp Gemma, +7.4 pp Qwen; the confirmed preregistered endpoint), justify themselves at greater length, and use fewer explicit uncertainty words in forced self-critiques (both exploratory). A fourth effect reverses sign: the same operation makes Gemma-abliterated less confident and Qwen-abliterated more (family CIs non-overlapping) - one weight surgery, opposite shifts in expressed confidence. Capability covariates rule out instruction-following degradation as the driver, and no arm shows economic skill: the apparent edge of abliterated arms is regime beta, not alpha. Our provenance audit also caught two independent contamination channels - a mismatched-quantizer pilot pair and a stale community chat template that silently mangled the rendered prompt - suggesting toolchain artifacts are the rule in studies of community-modified checkpoints. Whoever deploys an "uncensored" model as an agent is deploying a measurably different decision-maker, not the base model minus refusals.

AI 总结：针对高度非线性混沌动力系统建模难题，提出模糊谱区域分解（fSRD）方法，通过多个算子估计有限库普曼表示，实现数据自适应框架组装局部不变嵌入，在多系统和数据条件下展现出强预测性、可解释性和表现力。

链接：https://arxiv.org/abs/2607.17990

作者：Charles Bokor, Mark Cary, Denise Morrey, Fabrizio Bonatesta

英文摘要：Highly nonlinear chaotic dynamical systems remain difficult to model due to fundamental trade-offs between complexity, expressivity, and data efficiency. Modern machine learning methods achieve strong predictive performance but often rely on a-priori system knowledge or curated data with limited interpretability. Koopman operator theory offers a promising direction via linear representation in an infinite-dimensional observable space. However, many data-driven Koopman methods seek globally valid operators for which useful finite-dimensional spectral embeddings remain difficult to identify under these constraints. To overcome associated limitations, we introduce Fuzzy Spectral Region Decomposition (fSRD), a fully automated learning framework for estimating finite Koopman representation via multiple operators. The proposed method realizes a data-adaptive framework for assembling locally invariant embeddings, termed Invariant Decomposition. fSRD achieves highly accurate linear reconstructions of nonlinear systems while learning finite-dimensional representations of their induced evolution operators, bridging interpretable operator-theoretic models with expressive data-driven sequence learning. These embeddings are adaptively constructed via a global fuzzy tree model, drawing inspiration from fuzzy neural architectures to learn the induced dynamics while prioritizing parsimonious solutions. Empirical results across canonical chaotic systems (e.g., Lorenz and Duffing) and high-dimensional real-world data demonstrate strong predictive accuracy, interpretability, and robust expressivity across data-rich and data-limited regimes, highlighting the method's generality.

AI 总结：研究边缘设备上低延迟脑电图分类问题，提出可微逻辑门网络Diff-Logic，通过实验将其与MLP、BNN比较，结果表明Diff-Logic在痴呆筛查中表现优，推理时间稳定，确立其为资源受限脑机接口实用范式。

链接：https://arxiv.org/abs/2607.18149

机构：The University of Winnipeg（温尼伯大学）； University of Manitoba（曼尼托巴大学）； University of Calgary（卡尔加里大学）

作者：Shyamal Y. Dharia, Stephen D. Smith, Camilo E. Valderrama

英文摘要：Real-time EEG classification on edge devices is bottlenecked by the floating-point arithmetic of conventional neural networks. We investigated Differentiable Logic Gate Networks (Diff-Logic) as a hardware-native alternative that compiles models into pure Boolean circuits executable via bitwise CPU operations. Through rigorous iso-parameter experiments across four EEG datasets spanning two classification tasks, binary dementia detection and 3-class emotion recognition, we compared Diff-Logic against matched-capacity Multi-Layer Perceptron (MLP) and Binarized Neural Network (BNN) baselines at four complexity tiers (50k-500k parameters). On dementia screening, Diff-Logic achieved 80.2% Macro F1, outperforming the MLP baseline by 6.8%. On emotion recognition, the MLP retained a moderate performance advantage but incurred a 2.3$\times$ higher latency and 14$\times$ larger model size when deployed on a power-constrained (7W) Nvidia Jetson Orin Nano CPU (Single-core). Critically, Diff-Logic inference time remained nearly constant across a 10$\times$ increase in model scale, achieving a peak speedup of 2.9$\times$ over MLPs at the largest complexity tier. Our results establish logic-based neural architectures as a practical paradigm for resource-constrained brain-computer interfaces, achieving competitive or superior performance while natively satisfying the latency and memory constraints of portable edge deployment. Code is available on GitHub: this https URL

AI 总结：研究 PPO 和 GRPO 中裁剪代理目标导数突变问题，提出用输出重置（OR）规则优化。通过对比实验，在广义优势估计下 PPO-OR 有更高奖励模型得分，组相对优势下 GRPO-OR 虽平均得分未升但差异更小，OR 改变了优化行为但奖励效果有别。

链接：https://arxiv.org/abs/2607.18163

机构：Quantiphi Inc（昆蒂菲公司）； Self Machines Inc（自机器公司）； The University of Texas at Arlington（德克萨斯大学阿灵顿分校）

作者： Chinmay Rane, Kanishka Tyagi, Michael Manry

英文摘要：PPO and the GRPO baseline studied here use clipped surrogate objectives whose favorable-direction saturation introduces an abrupt change in the scalar objective's derivative. We ask whether Output Reset (OR), a smooth one-sided saturation rule, offers a useful alternative for large language model post-training. PPO-OR and GRPO-OR replace the clipped policy term with an OR squared-margin loss in rollout-relative token log-ratio space; the advantage sign determines the update direction, and a token contributes zero direct OR residual after crossing the favorable margin. We compare PPO-clip with PPO-OR under generalized advantage estimation (GAE), and GRPO with GRPO-OR under group-relative advantages, using \texttt{Llama-3.2-1B-Instruct} on Anthropic \texttt{hh-rlhf} with one shared reward model and three seeds per method. Under GAE, PPO-OR has a mean final training-time reward-model score $0.305$ higher than PPO-clip, with a larger observed across-seed spread. Under group-relative advantages, GRPO-OR does not have a higher mean score, but shows a smaller observed spread, a near-zero terminal OR residual, and a declining overshoot fraction, while the matched GRPO clipped-objective trace remains variable. Both group-relative methods exhibit substantially larger rollout-to-current log-ratio displacement than the GAE methods, and OR does not consistently reduce it. Thus, OR changes optimization behavior in both matched comparisons, but the observed reward effect differs between them. At $G=2$, the GRPO-OR diagnostics do not translate into a reward-score gain. Whether larger groups change this outcome remains open. The reported scores are training-time reward-model measurements, not held-out human-preference performance.

AI 总结：研究神经网络激活可解释性方法，提出HARP方法，即让语言模型智能体利用激活向量数据库及工具，通过迭代查询、形成并验证假设来实现无需训练的可解释性，该方法在多方面表现出色，还凸显现有训练方法局限并推动新基准测试。

链接：https://arxiv.org/abs/2607.16448

机构：University of Maryland（马里兰大学）

作者：Sriram Balasubramanian, Soheil Feizi

英文摘要：Interpretability methods for neural network activations span a wide cost spectrum, from cheap, training-free techniques (such as linear probes, PCA, SVD) to more expensive training-based ones (such as SAEs and activation oracles). Training-based methods are typically more powerful, in part because they leverage large activation datasets during training. This raises a natural question - do they actually surface insights that go beyond what is recoverable from the training dataset itself? To address this, we equip an LLM agent with a vector database of activations paired with their textual contexts, along with tools for manipulating activations - projecting out directions in latent space, computing activation differences and averages. The agent iteratively queries the database, forms hypotheses from the retrieved samples, and validates them by constructing linear probes. We call this method HARP, for Hypothesis-driven Agentic Retrieval and Probing. Despite not involving any training, HARP outperforms both activation oracles and SAE-based agents on concept discovery, concept detection, model steering, and secret elicitation. The training-free design also makes HARP substantially cheaper and more flexible: new datasets can be indexed on demand whenever existing ones prove insufficient. More broadly, our results suggest that current training-based methods do not yet extract insights beyond their training data, and motivate benchmarks that explicitly require interpretability methods to demonstrate such insights. We release our code at this https URL

AI 总结：研究针对电子健康记录中早期脓毒症预测面临的问题，比较了JEPA等四种建模范式及原始特征基线，最佳模型在发病时接近SupMix基准且减少生物标志物使用，一级管道有显著提升，微调后的VICReg编码器特征持久性最佳。

链接：https://arxiv.org/abs/2607.16681

作者：Umair bin Mansoor, Munaf Rashid, Roomi Naqvi

英文摘要：Early sepsis prediction from electronic health records is challenged by irregular sampling, high missingness, and class imbalance. We systematically compare four modeling paradigms -- self-supervised Joint Embedding Predictive Architecture (JEPA) via masked latent prediction, self-supervised VICReg (variance-invariance-covariance regularization) with two-view augmentation, semi-supervised fine-tuning of a VICReg-pretrained encoder, and supervised Temporal Convolutional Network (TCN) -- alongside raw-feature baselines. All models share a common preprocessing pipeline of hourly binning with forward-fill imputation applied to 7 biomarkers selected via sparsity analysis from the MIMIC-III dataset. Our best model (JEPA + XGBoost + mean pooling) achieves AUPRC 0.636 at the time of onset (H0), approaching the SupMix benchmark (0.667) while using 83\% fewer biomarkers. The Tier 1 pipeline -- VICReg pretraining followed by semi-supervised fine-tuning and XGBoost -- achieves AUPRC 0.510 at H0, a 3.1$\times$ improvement over the raw-feature baseline (0.165) and a 7.6\% improvement over the end-to-end supervised TCN (0.474). Crucially, the fine-tuned VICReg encoder exhibits the most temporally persistent representations, degrading only 16.8\% from H0 to H10 compared to 47.5\% for supervised TCN and 65.3\% for JEPA, demonstrating that self-supervised pretraining with task-aware fine-tuning yields features that are both sharp near onset and robust across prediction horizons.

AI 总结：研究针对多数节点表示学习方法依赖特定数据集训练的问题，提出Node4All，基于通道图变换器和自监督学习构建，能跨任意图数据集泛化，在节点分类任务中表现出色，实现可复用性且实践效果好。

链接：https://arxiv.org/abs/2607.17272

机构：KAIST（韩国科学技术院）； Seoul National University（首尔国立大学）

作者：Dooho Lee, Jaemin Yoo

英文摘要： Node representation learning has advanced rapidly, yet most existing methods rely on per-dataset training and hyperparameter tuning. This dataset-specific optimization comes from the difficulty of designing reusable graph models that generalize across diverse graph datasets. In this work, we introduce Node4All, a node representation learner applicable to arbitrary graph datasets without any dataset-specific optimization. Node4All is built on two complementary ideas. At the architectural level, we introduce the Channel Graph Transformer (CGT), which enables a single fixed parameterization to process arbitrary graph datasets. At the learning level, we propose a self-supervised learning based on a series of synthetic graphs. Together, these components enable generalization beyond individual datasets, which is infeasible with existing architectures and learning frameworks. We extensively evaluate Node4All on node classification across 25 benchmarks against 21 baselines, covering both supervised and self-supervised methods. Despite all baselines being trained and optimized for each dataset, a single Node4All, applied uniformly across the datasets, achieves a competitive ranking of 5th among 21 baselines. Moreover, Node4All supports one-shot and in-context learning with an appropriate predictor and outperforms recent graph foundation models (GFMs) in these settings. These results demonstrate that Node4All not only achieves reusability across arbitrary graph datasets, but also remains an effective solution in practice. Code and model checkpoints are available in this https URL.

AI 总结：研究稀疏自编码器中哪些读出能在稀疏压缩中幸存的问题，核心方法是将最优岭预测算子间的矩阵值失真与重建损失结合训练解码器保留SAEs，主要贡献为得出重建质量不能决定读出幸存情况等结论。

链接：https://arxiv.org/abs/2607.17425

机构：University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）

作者：Aniket Deshpande

英文摘要：Sparse autoencoders (SAEs) compress model activations into sparse codes, but equal reconstruction error and sparsity can preserve different linearly decodable signals. We formalize this ambiguity as a matrix-valued distortion between optimal ridge-prediction operators and train decoder-preserving SAEs by combining this distortion with reconstruction loss. In a rank relaxation, an isotropic task prior saturates per-mode omission costs without changing PCA's ordering, whereas a structured prior can change which modes are retained. A controlled sparse experiment shows that a declared prior protects held-out combinations from its task subspace. On GPT-2 small block 8, DPSAE reduces held-out decoder distortion by 10.6--11.4% across three paired runs while matching reconstruction NMSE. The same checkpoints pass an average natural-text output-KL noninferiority test, but one matched Pythia pair shows no improvement in probes restricted to a few sparse features. These results show that reconstruction quality does not determine which refitted linear readouts survive sparse compression, and that readout preservation is distinct from learning cleaner benchmark concepts or preserving every frozen-model behavior.

AI 总结：研究多模态对比学习从成对扩展到高阶对齐的挑战，确定编码器雅可比条件是关键因素，引入几何保留编码器，通过正则化调节雅可比，实验证明改善其条件可提升检索和线性探针性能，表明多模态对比学习还取决于编码器几何和优化属性。

链接：https://arxiv.org/abs/2607.17673

机构：Berlin Institute of Health, Charité - Universitätsmedizin Berlin（柏林健康研究院，柏林夏里特大学医学中心）； Department of Mathematics and Computer Science, Freie Universität Berlin（柏林自由大学数学与计算机科学系）； Intelligent Medicine Institute, Fudan University（复旦大学智能医学研究院）

作者：Tillmann Rheude, Roland Eils, Benjamin Wild

英文摘要：Contrastive learning is increasingly moving toward settings with three or more modalities instead of image-text pairs. Yet, extending models from pairwise to higher-order multimodal alignment can introduce optimization and representation challenges. We identify encoder Jacobian conditioning as a key factor in trimodal contrastive learning: poorly conditioned encoders exhibit collapsing or amplified singular-value spectra, leading to exploding Jacobian condition numbers and degraded multimodal alignment. We introduce geometry-preserving encoders (GPEs) by directly conditioning the Jacobian through regularization and demonstrating that simple modifications like LeakyReLU activations and residual paths recover these geometric benefits. Across a synthetic benchmark and four real-world datasets including missing modalities, improving Jacobian conditioning boosts retrieval and linear probe performance across multiple contrastive objectives, whereas expressive objectives yield little benefit in linear probes. More broadly, our results show that multimodal contrastive learning depends not only on objective expressivity, but also on the geometric and optimization properties of the underlying encoders.

AI 总结：研究针对均方根归一化问题，提出MRSNorm方法，通过将通道配对成相量改变缩放范式，约束激活在相量流形，共享权重减半参数，经分析和实验验证其能确保梯度均匀性、提供结构稳定性及防止数值爆炸，推动向基于相量的深度表示学习转变。

链接：https://arxiv.org/abs/2607.17822

机构：College of Pharmacy, Chungnam National University（忠南国立大学药学院）； Department of Computer Science & Engineering, Chungnam National University（忠南国立大学计算机科学与工程系）

作者：Sungwoo Goo, Hwi-yeol Yun, Sangkeun Jung

英文摘要：While Root Mean Square Normalization has become the de facto standard for accelerating modern sequence models, its reliance on the quadratic accumulation of independent scalars ($\sum x^2$) inherently triggers outlier-induced numerical instability, gradient starvation, and anisotropic phase distortion. We introduce Mean Root Square Normalization (MRSNorm). By structurally pairing channels into 2D phasors, MRSNorm mathematically inverts the traditional scaling paradigm: it computes the localized $L_2$ magnitudes (Root Square) before aggregating them via a global $L_1$ average (Mean). This operational inversion strictly constrains activations to a phasor manifold, preserving conformal invariance. By sharing a single affine weight across phasor components, MRSNorm halves the total number of learnable parameters, proving that unconstrained spatial scaling in standard norms is a harmful redundancy. We analytically demonstrate that this geometric constraint yields a built-in, trigonometric gradient clipper governed by the Pythagorean identity, unconditionally equalizing the local gradient norm to ensure Gradient Homogeneity. Empirical evaluations on a ResNet with CIFAR-100 show that despite halved parameters, MRSNorm provides critical structural stability under rigorous stress tests. Under extreme hyperparameter settings where standard normalizations suffer from gradient divergence, MRSNorm successfully prevents numerical explosion and secures stable optimization trajectories. Our findings propose a fundamental paradigm shift toward phasor-based deep representation learning. The implementation of MRSNorm is available at Appendix C.

AI 总结：本文针对投资组合优化中的约束多目标优化问题，提出RL-NSGA-II-GRC方法，结合强化学习智能体与灰色关联系数，通过自适应控制参数及设计综合指标引导搜索，在基准测试和纳斯达克案例中提升收敛性，得到密集前沿及有效投资组合。

链接：https://arxiv.org/abs/2607.16194

作者：Zhiyuan Wang, Qinxu Ding, Ding Ding, Siying Zhu, Jing Ren, Yue Wang, Chong Hui Tan

英文摘要：In modern financial markets, decision-makers increasingly rely on quantitative methods to navigate complex trade-offs among multiple, often conflicting objectives. This paper addresses constrained multi-objective optimization (MOO) with an application to portfolio optimization for minimizing risk and maximizing return. To address existing gaps, we propose a novel reinforcement learning (RL)-guided non-dominated sorting genetic algorithm II (NSGA-II) enhanced with gray relational coefficients (GRC), termed RL-NSGA-II-GRC, which combines an RL agent controller and GRC-based selection to improve convergence and diversity of Pareto fronts. The agent adapts evolutionary parameters online using metrics of hypervolume, feasibility, and diversity, while the GRC tournament operator ranks parents via a unified score considering dominance rank, crowding distance, and proximity to ideal reference. We evaluate the framework on the Kursawe and CONSTR benchmarks and a NASDAQ portfolio application. On the benchmarks, RL-NSGA-II-GRC achieves convergence improvements of about 5.8% and 4.4% over NSGA-II, while preserving well-distributed non-dominated solutions. In the portfolio application, it produces a smooth, densely populated efficient frontier supporting identification of the maximum Sharpe ratio portfolio (annualized Sharpe =1.92) and utility-optimal portfolios for different risk-aversion levels. The main contributions are three-fold: 1) we propose an RL-NSGA-II-GRC method integrating an RL agent into the evolutionary framework to adaptively control parameters via generational feedback; 2) we design a GRC-enhanced binary tournament operator providing a comprehensive indicator to guide the search toward the Pareto front; 3) we demonstrate, on benchmark MOO and a NASDAQ case study, that the method delivers improved convergence and well-populated frontiers supporting actionable insights.

AI 总结：研究多轮证据阅读语言模型代理训练不稳定问题，提出CIGPO方法，通过方差注入策略，利用冻结参考模型对数似然边际增加作逐轮信号，避免奖励方差崩溃，在HotpotQA实验中取得更好效果。

链接：https://arxiv.org/abs/2607.16244

机构：Harbin Institute of Technology（哈尔滨工业大学）

作者：Hao Dou

英文摘要：Training multi-turn evidence-reading agents with outcome-only reinforcement learning is unstable because intermediate turns receive little direct credit. In HotpotQA experiments with Qwen2.5-3B-Instruct, GRPO initially improves (standard F1 0.430) but subsequently collapses to 100% format-violating outputs. Training-log diagnosis reveals a zero-advantage lock-in mechanism: all sampled trajectories receive the minimum format penalty (-2.0), group-relative advantages vanish, and the policy-gradient loss becomes zero--an optimization deadlock. We propose a variance-injection strategy: by assigning per-turn rewards to intermediate evidence-reading turns, we prevent the group reward distribution from collapsing to a single value--preserving the variation that GRPO's group-relative advantage requires. Contextual Information-Gain Policy Optimization (CIGPO) implements this strategy using the marginal increase in the frozen reference model's log-likelihood of the ground-truth answer as the per-turn signal. With separate normalization of IG and F1 rewards and an IG-weight curriculum, CIGPO reaches a standard F1 of 0.518 on HotpotQA at the 3B scale (from 0.252 base; +105%), compared with 0.430 for the best GRPO checkpoint and 0.000 for the final GRPO checkpoint. CIGPO maintains meaningful reward variance and avoids zero-advantage lock-in throughout training. These results identify reward-variance collapse as a concrete failure mode of outcome-only GRPO and show that turn-level IG rewards can prevent it in this HotpotQA setting.

AI 总结：研究长期语言智能体训练中现有强化学习方法面临的挑战，提出事后诸葛亮策略优化方法，通过投影到意图空间提取低方差学习信号，聚合相似状态和行动，理论和实证证明可稳定提升策略性能。

链接：https://arxiv.org/abs/2607.16257

作者：Zishang Jiang, Tingyun Li, Jinyi Han, Xinyi Wang, Sihang Jiang, Yizhou Ying, Xiaojun Meng, Jiansheng Wei, Jiaqing Liang, Yanghua Xiao

英文摘要：Reinforcement learning (RL) has become a widely adopted technique for improving large language models (LLMs) on complex tasks. Despite this progress, existing RL methods still face challenges in training agents with longer-horizon interactions. One major bottleneck is distinguishing the contribution of different actions in long-horizon interaction, leading to high optimization variance. To address this, we introduce a novel policy gradient method, Hindsight Policy Optimization (HPO), that projects both the current policy distribution and the hindsight distribution into an intent space and extracts low-variance learning signals from the Wasserstein distance between them. We theoretically and empirically show that aggregating semantically similar states and actions in the intent space yields a bounded-variance estimator and improves policy performance stably. Our code is available online.

AI 总结：研究针对零售需求预测难题，提出预测-校正框架，运用少样本连续上下文博弈校正策略等，经实验在多需求模式下显著降低误差、提高RMSE并降低库存成本，证明在线预测校正能连接离线需求学习与实时零售决策。

链接：https://arxiv.org/abs/2607.16354

机构：Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））； Massachusetts Institute of Technology（麻省理工学院）

作者：Zhiwei Lei, Benedict Jun Ma, Ilya Jackson

英文摘要： Retail demand forecasting remains difficult when demand shifts faster than static forecasting models can be retrained, especially in early demand cycles where newly observed labels are sparse. To address this, this study aims to improve adaptive retail forecasting by proposing a predict-then-correct (PtC) framework that retains a first-stage machine learning (ML) forecast and applies a few-shot continuous contextual bandit correction policy with similar-SKUs augmentation and top-p masked updating. Across Walmart retail data and an exclusive beverage dataset, PtC delivers statistically significant reductions in MAPE, MAE, and RMSE across stable & high volume, stable & low volume, and erratic & intermittent demand patterns, improves average RMSE by 9.52% over the ML-only baseline in the ablation study, and yields lower inventory costs than base-stock, proximal policy optimization, and soft actor-critic policies under the tested lead-time settings. These findings show that online forecast correction can bridge offline demand learning and real-time retail decision-making by adapting to sparse feedback without fully retraining the base forecasting model.

AI 总结：研究多智能体强化学习中团队平均奖励下，学习表示上奖励归因的特征。提出EffRank/$n$和$D_\text{act}$指标，在SMACv2测试发现观测解释几何特征，奖励归因体现在行为，且指标开销低。

链接：https://arxiv.org/abs/2607.16524

作者：Tasha Pais, Richard Higgins

英文摘要：Cooperative multi-agent RL systems routinely use team-averaged rewards, a feedback-attribution choice that gives each agent the team outcome regardless of its individual contribution. We ask whether this leaves a measurable signature, geometric or behavioral, on learned representations. We propose EffRank/$n$ (effective rank normalized by agent count) and $D_\text{act}$ (mean pairwise KL divergence between agents' action distributions) as low-overhead diagnostics for reward-attribution effects, then test them on competent MAPPO agents in SMACv2 \texttt{protoss\_5\_vs\_5}, where unit type is encoded in the observation. In an observation $\times$ reward-attribution comparison (unit type observed vs.\ masked; individual damage-contribution reward vs.\ shared team reward), geometry follows observation rather than reward. With unit type observed, shared and individual rewards have similar EffRank/$n$ ($0.31{\pm}0.03$ vs.\ $0.29{\pm}0.02$) and probe accuracy ($0.75{\pm}0.05$ vs.\ $0.73{\pm}0.05$, both $\gg 1/3$ chance), while $D_\text{act}$ leans higher under individual rewards ($1.23{\pm}0.06$ vs.\ $1.07{\pm}0.20$). Masking unit type cuts the above-chance probe signal by more than half, to $0.49$ in both reward arms. In short: individually rewarded agents are competent and separable by role, but on SMACv2 the observation explains the geometry and reward attribution shows up mainly in behavior. Thus geometric diagnostics must control for observed role information and test persistent roles that are not directly observed. EffRank/$n$ and $D_\text{act}$ add $

AI 总结：介绍用于可泛化现实世界强化学习的Building2Building基准测试，它基于EnergyPlus构建，有多样建筑配置，定义针对RL关键挑战的基准任务，能系统研究泛化与转移，推进RL研究并对HVAC控制有社会意义。

链接：https://arxiv.org/abs/2607.16534

作者：Vincent Taboga, Justin Veilleux, Doseok Jang, Anushree Rankawat, Pierre-Luc Bacon

英文摘要：Reinforcement learning (RL) has achieved strong results in control, yet learned policies remain brittle to changes in dynamics, action spaces, observation spaces, or goals, a critical limitation for real-world deployment. Existing benchmarks offer limited diversity and complexity, making it difficult to rigorously study transfer, multi-task learning, and meta-learning in RL. We introduce Building2Building (B2B), a large-scale suite of realistic Heating, Ventilation, and Air Conditioning (HVAC) control environments built on EnergyPlus, a state-of-the-art building simulator. B2B is fully compatible with the Gymnasium interface and features a parametric building generator, enabling the systematic generation of diverse building configurations with heterogeneous observation and action spaces. Based on this suite, we define benchmark tasks targeting key open challenges in RL, including goal adaptation, dynamics adaptation, action-space shifts, and cross-domain transfer. By providing a large-scale, diverse, and physically grounded testbed with standardized evaluation protocols, B2B enables systematic investigation of generalization and transfer in continuous control. Beyond advancing research on generalization in RL, this new benchmark also carries significant societal implications by enabling improved HVAC control at scale, one of the most energy-intensive systems in buildings.

AI 总结：研究探讨 RLxF 中学习信号应源于世界反馈，在安全模型控制中实例化并提炼原则。通过实验表明基于动力学的不确定性惩罚会增加碰撞率，用世界反馈信号可降低碰撞率，提取原则并指出其适用于多种相关方法。

链接：https://arxiv.org/abs/2607.16591

作者：Zhaohui Wang

英文摘要：The RLxF programme argues that learning signals should come from world feedback rather than from internal model proxies. We instantiate this position in safe model-based control and distil it into three concrete design principles. Empirically, across four world-model architectures spanning a 2x MSE range, MPC planning is statistically equivalent (TOST, n=200), and dynamics-based uncertainty penalties increase collision rates from 26% to 34%: the standard MBRL safety proxy is anti-correlated with safety in this regime. Replacing the model-internal proxy with three world-feedback signals (a sensor-derived margin via minimum lidar, a temporal signal via time-to-collision, and an outcome-supervised feedback model g_psi trained on prior collision labels, structurally analogous to outcome-trained reward models in RLHF) reduces collisions to 1-14% without retraining the world model or the planner. The mechanism is structural: model uncertainty has support over state-prediction space, whereas task risk has support over constraint boundaries, with empirical correlation r < 0.15. From this we extract three RLxF principles (ground risk in world outcomes, validate proxies before deployment, and substitute outcome-trained feedback models when direct world signals are unavailable) and argue they apply equally to model-based control and to verifier-based or RLHF approaches in LLM alignment.

AI 总结：研究在含混合不确定性的环境中无监督强化学习的内在动机问题，提出基于无偏好预期自由能目标新颖性贡献的单一内在奖励，通过参数信息增益驱动探索，给出相关技术细节及成果，使智能体在不同状态区域表现良好。

链接：https://arxiv.org/abs/2607.16858

机构：Coovally（库瓦利）； ICREA（加泰罗尼亚研究与高等研究院）； Distributed Systems Group, TU Vienna（维也纳工业大学分布式系统组）

作者：Alireza Furutanpey, Schahram Dustdar

英文摘要：Across environments with mixed sources of uncertainty, unsupervised reinforcement learning requires intrinsic motivation that does not precommit to a particular direction of surprise. Surprise minimization is scoped by design to ``unstable'' environments. Prediction-error curiosity rewards total expected surprise, including irreducible noise. Bandit or mixture switching between surprise-minimizing and surprise-maximizing rewards reintroduces non-stationarity by construction. We propose a single intrinsic reward, stationary within each window, derived from the novelty contribution of a preference-free Expected Free Energy objective, expressed in reward-maximization form. Our claim is that parameter information gain, the expected surprise of the next state minus its irreducible part, is the appropriate intrinsic signal in both high-entropy and low-entropy components of the state space. Maximizing it seeks exactly the surprise the model can explain away. In regions of unresolved dynamics, this epistemic term drives exploration. As dynamics become resolved, the epistemic term vanishes, while an aleatoric penalty favors lower-variance transitions, all without fitting an explicit next-state predictor. A pseudocount supplies epistemic value, a probe-based penalty captures aleatoric variance, and a short-horizon gate protects informative successors. A window-based freeze of all reward-defining objects yields a stationary Bellman operator, explicit bounds on learning targets, and a conditional uniform-concentration result for the nonparametric estimators under mixing, smoothness, bandwidth, and capacity assumptions. In active-inference terms, the agent is preference-free where novelty is retained, standard likelihood ambiguity vanishes under full observability, a nonstandard transition-entropy penalty is added, and surprise minimization emerges in resolved regions of the state space.

AI 总结：研究针对膀胱癌复发性疾病治疗需个性化决策的问题，提出集成预测状态转换建模、马尔可夫决策过程和深度Q网络强化学习环境的框架，能动态制定治疗计划，经评估效果良好，凸显其作为个性化膀胱癌治疗规划决策支持框架的潜力。

链接：https://arxiv.org/abs/2607.16916

作者：Divyansh Chawla, Anshu Garg, Isshaan Singh

英文摘要：Bladder cancer treatment requires personalized and adaptive decision-making, particularly for recurrent disease, where treatment effectiveness changes across successive clinical episodes. Conventional clinical decision support systems typically rely on static treatment guidelines or single-step predictive models, limiting their ability to capture disease progression over time. This paper presents a recurrent patient state-transition simulation framework for bladder cancer treatment planning that integrates predictive state-transition modeling with a Markov Decision Process (MDP) and a Deep Q-Network (DQN) reinforcement learning environment. The predictive module estimates changes in tumor characteristics following treatment, while the reinforcement learning agent sequentially optimizes treatment decisions by interacting with simulated patient trajectories. This framework enables dynamic, patient-specific treatment planning by continuously adapting recommendations to evolving clinical states. It also generates interpretable treatment trajectories and detailed simulation logs to improve transparency and support clinical decision-making. The proposed framework was evaluated against existing reinforcement learning-based treatment planning approaches. It achieved a cumulative reward of 63,918.87, an average training loss per episode of 0.0056, and a policy improvement score of 6.62%, demonstrating effective sequential learning and robust treatment optimization in a simulated recurrent treatment environment. These findings highlight the potential of recurrent patient state-transition simulation with reinforcement learning as a flexible decision-support framework for personalized bladder cancer treatment planning and AI-assisted precision oncology.

AI 总结：针对周期性非平稳多臂老虎机问题，提出周期性自展汤普森采样（PBTS），通过同步信念重置、嵌入自展探索阶段克服传统TS缺陷。实验表明PBTS能显著降低累积遗憾，还阐述了其实际部署潜力与局限及未来研究方向。

链接：https://arxiv.org/abs/2607.16986

作者：Boning Shao

英文摘要：This paper introduces Periodic Bootstrap Thompson Sampling (PBTS), an innovative extension of the classic Thompson Sampling (TS) algorithm tailored for bandit problems with periodic non-stationarity. Conventional TS accumulates all past observations, leading to biased posteriors when reward distributions cycle over time. PBTS overcomes this by synchronizing belief resets with known or inferred period intervals and embedding structured bootstrap exploration phases, effectively purging obsolete data while preserving uncertainty estimates. PBTS is tested in artificially constructed environments, which include skewed and balanced reward distributions, along with different bootstrap proportions and misaligned periodic intervals. Results indicate that PBTS generally achieves statistically significant reductions in cumulative regret against traditional TS in periodic non-stationary environments. Subsequent discussion further articulates the potential of PBTS's real-world deployment. The study mentions limitations like extreme periodic misalignment and proposes future research such as self-adjusting cycle-recognition. With memory reset and bootstrap phase, PBTS introduces a novel approach to optimizing bandit algorithms in periodic reward contexts.

AI 总结：研究信用分配问题，提出基于因果理论的反事实沙普利信用分配框架，通过反事实沙普利值分配功劳责任，推导有效估计器实现$\phi$-PPO方法，结合优先轨迹重放，在复杂环境中有卓越样本效率，能精确对齐奖励真实原因。

链接：https://arxiv.org/abs/2607.16999

作者：Mingxuan Li, Kaizhan-Lee, Elias Bareinboim

英文摘要： The Credit Assignment Problem (CAP) is fundamental to developing efficient and explainable Reinforcement Learning (RL) agents. Existing frameworks, whether relying on temporal contiguity or hindsight-conditioned reward reweighting, frequently fail to attribute properly between an agent's policy (skill) and environmental stochasticity (luck). A principled approach to CAP must isolate the true causal drivers of observed outcomes from spurious correlations and environmental randomness. We introduce Counterfactual Shapley Credit Assignment, a novel framework grounded in causal theory that attributes credit and blame via the Counterfactual Shapley Value ($\phi$-value). By redistributing environmental rewards, $\phi$-values enhance temporal credit assignment across three critical dimensions: sparse causality, high stochasticity, and delayed rewards, all while preserving the optimal policy. We derive a consistent estimator that computes $\phi$-values efficiently, enabling a new class of policy gradient methods, $\phi$-PPO, combined with Prioritized Trajectory Replay (PTR). Empirical results demonstrate that $\phi$-values align precisely to the ground truth causes of task rewards with superior sample efficiency in challenging environments where prior state-of-the-art methods fail to converge.

AI 总结：研究在模仿学习观测不匹配及有未观测混杂因素时的问题，提出因果软Q模仿学习和因果逆软Q学习算法，结合因果调整框架与逆强化学习目标，在长视任务中显著优于现有CIL算法。

链接：https://arxiv.org/abs/2607.17003

作者：Eylam Tagor, Mingxuan Li, Elias Bareinboim

英文摘要：Imitation learning enables learning a policy in an unknown environment with a latent reward signal using expert demonstrations, but it struggles when the imitator's and expert's observations are mismatched and unobserved confounders are present in expert demonstrations. By identifying appropriate adjustment sets via the sequential $\pi$-backdoor criterion, causal imitation learning (CIL) provides a framework for approximating the expert's policy from confounded data. However, existing CIL methods, Causal Behavioral Cloning (Causal BC) and Causal Generative Adversarial Imitation Learning (Causal GAIL), are designed for short-horizon, low-dimensional settings. When applied to continuous control tasks with long horizons and high-dimensional state-action spaces, these methods exhibit poor performance: Causal BC suffers from compounding errors, Causal GAIL is unstable and sample-inefficient, and sequential $\pi$-backdoor adjustment becomes impractical. We introduce Causal Soft Q Imitation Learning (SQIL) and Causal Inverse soft-Q Learning (IQ-Learn), two off-policy causal imitation learning algorithms that combine the causal adjustment framework with state-of-the-art inverse reinforcement learning objectives. Both algorithms operate on causally-adjusted state representations produced by an efficient approximation of the sequential $\pi$-backdoor criterion, exploiting the causal structure of continuous control environments to reduce the full-horizon adjustment to a fixed-size sliding window. We evaluate all methods in a suite of confounded environments and find that Causal SQIL and Causal IQ-Learn substantially outperform prior CIL algorithms on long-horizon tasks, sometimes surpassing the expert, whereas all causally unaware imitation methods fail to learn meaningful behavior.

AI 总结：研究针对智能体强化学习中展开成为系统瓶颈的问题，提出工作负载感知展开系统WAR，通过联合优化解码和调度加速同步智能体RL，在不同负载下提高吞吐量，消除主要展开瓶颈，为长上下文智能体训练提供实用路径。

链接：https://arxiv.org/abs/2607.17299

作者：Ryan Xu, Atlas Zhao, David Bao, Frank Du

英文摘要：Long-horizon rollout generation has become the dominant systems bottleneck in agentic reinforcement learning (RL). As agents interact with environments over many turns, trajectories rapidly grow to tens of thousands of tokens, making synchronous RL training increasingly constrained by rollout. We propose WAR, a workload-aware rollout system that substantially accelerates synchronous agentic RL by jointly optimizing decoding and scheduling. WAR is built on a key observation: the optimal rollout optimization strategy depends on runtime load: (1) Under low load, WAR enables model-free speculative decoding with SuffixDecoding, which reuses suffix patterns from previously completed trajectories as speculative drafts for future rollouts. Unlike model-based drafters, SuffixDecoding introduces no additional draft model and avoids GPU contention with rollout generation. (2) Under high load, where saturated batched decoding leaves limited room for speculative speedup, WAR shifts the optimization focus to cache-aware scheduling. A global scheduler places requests across rollout replicas based on cache locality, trajectory progress and server load, reducing redundant KV-cache recomputation and mitigating load imbalance. By combining decoding-level suffix reuse with system-level rollout scheduling, WAR delivers robust throughput improvements across workload regimes without changing the underlying RL algorithm. WAR improves long-context agentic rollout throughput by 1.4x under low load and up to 1.6x under high load. These results show that WAR removes a major rollout bottleneck in synchronous agentic RL and provides a practical path toward scalable long-context agent training.

AI 总结：研究强化学习中softmax策略，通过区分两种随机性化解其与MDP奖励结构独立性公理的矛盾，施加IIA和单调性确定玻尔兹曼策略等，得出RL相关结果，综合多领域线索对IIA用于智能体设计进行规范评估。

链接：https://arxiv.org/abs/2607.17316

作者：Silviu Pitis

英文摘要：The softmax policy $\pi(a \mid s) \propto \exp(\beta Q(s,a))$ is the default model of stochastic choice in reinforcement learning (RL). Various justifications based on robustness, exploration, and optimization have been offered in the RL literature, but none uniquely derives the softmax form from first principles. This leaves a basic tension unresolved: the entropy bonus in the soft Bellman equation violates the Independence axiom that underwrites the Markov decision process (MDP) reward structure. We dissolve this tension by distinguishing two kinds of randomness: chance and choice. By restricting von Neumann-Morgenstern (VNM) Independence to environmental lotteries over base prospects, we show that imposing independence of irrelevant alternatives (IIA) and monotonicity on the policy and value functions at choice nodes uniquely determines the Boltzmann policy, the entropy-regularized representation, and the soft Bellman equation. The choice between the soft and hard Bellman equations thus reduces to a design decision: whether the agent values its own ability to choose. We develop RL-specific consequences, including return monotonicity and convergence under generalized discounting, and synthesize the independent lines from economics and information theory that arrive at the same structure, offering a normative assessment of when IIA is appropriate for agent design.

AI 总结：针对自主系统感知模块吞吐量需求随场景变化的问题，提出基于强化学习的吞吐量自适应感知策略TAPAS，能在异构移动/边缘平台智能分配资源，在KITTI和nuScenes数据集上评估，吞吐量满足率高且节能效果显著。

链接：https://arxiv.org/abs/2607.17317

机构：University of Turku（图尔库大学）

作者：Aman Vyas, Vasista Kodumagulla, Zain Taufique, Pasi Liljeberg, Anil Kanduri

英文摘要：Autonomous systems rely on a perception module to navigate through dynamic environments. In real-world scenarios, the perception module's throughput requirements vary at runtime due to changes in scene complexity. However, existing perception strategies assume a fixed FPS and static model-to-cluster mapping, resulting in either over/under provision of throughput requirements or unnecessary energy consumption across diverse scenes. Addressing this challenge requires tightly coupled \textit{scene complexity awareness} to estimate an appropriate FPS target and \textit{dynamic model-to-cluster mapping} to deliver the required throughput at minimum energy. We propose a throughput-adaptive perception strategy for mobile/edge platforms, enabling intelligent runtime resource allocation based on varying FPS targets. We use Reinforcement Learning (RL) with RRM (Reward Reasoning Model) and a GRU (Gated Recurrent Unit) agent to orchestrate perception tasks across heterogeneous mobile/edge platforms. We evaluate TAPAS on Jetson Orin NX across KITTI and unseen nuScenes. On the \textit{KITTI} dataset's test sequences, TAPAS achieves 93-100% throughput met rate while saving energy by 76%. On the unseen \textit{nuScenes} dataset, TAPAS maintains 97% throughput met rate with 64% lower energy compared to \textit{SOTA} approaches, proving its robustness.

AI 总结：研究实际迁移约束下强化学习算法适用性，从实际效率和动态不匹配鲁棒性两维度评估。发现PPO虽样本效率低但训练快，域随机化对不同算法影响相似，强调评估算法不能仅看样本效率，还需考虑实际因素。

链接：https://arxiv.org/abs/2607.17326

作者：Hany Hamed, Abhishek Naik, Colin Bellinger, A. Rupam Mahmood

英文摘要：Transfer-oriented reinforcement learning requires evaluating algorithms along dimensions that go beyond standard sample efficiency. We focus on two dimensions: practical efficiency, which asks whether conclusions about algorithm suitability change under wall-clock rather than interaction-based budgets, and robustness under dynamics mismatch, which asks how different learning paradigms respond to variability in the training distribution induced by domain randomization. We provide two insights to reinforcement-learning practitioners. First, comparing the sample efficiency of different algorithms is often an insufficient criterion in transfer-oriented settings. The wall-clock time required to train a decent policy is an important consideration for practitioners, and we find that the sample-inefficient PPO algorithm can produce a performant policy faster than relatively more sample-efficient algorithms such as SAC and TD-MPC2, validating the common understanding of massively parallel training paradigms. Second, domain randomization can help different kinds of algorithms learn robust policies. In particular, although PPO, SAC, and TD-MPC2 represent different RL paradigms - on-policy, off-policy, and model-based learning and planning, respectively - we find that domain randomization affects all three algorithms in a similar way. To the best of our knowledge, this is the first controlled comparison of the effect of domain-randomization coverage on PPO, SAC, and TD-MPC2 under the same transfer protocol. Taken together, these two insights highlight the importance of evaluating RL algorithms not only by sample efficiency, but also by practical considerations such as training time and the algorithms' ability to produce usable policies.

AI 总结：研究针对淀粉样纤维配体对接面临的挑战，提出CORAL强化学习框架，通过纳入协同配体 - 配体堆叠能量及蛋白质 - 配体对接亲和力训练模型，引入评估集，实验证明其在姿态质量和结合亲和力相关性上优于现有基线。

链接：https://arxiv.org/abs/2607.17412

机构：Center of Excellence for Generative AI, KAUST（沙特阿卜杜拉国王科技大学生成式人工智能卓越中心）； MoE Key Lab of Artificial Intelligence, Shanghai Jiao Tong University（上海交通大学人工智能教育部重点实验室）； Stern Laboratory, Brigham and Women’s Hospital, Harvard University（哈佛大学布莱根妇女医院斯特恩实验室）

作者：Yasheng Sun, Bohan Li, Youqi Tao, Jürgen Schmidhuber

英文摘要：A hallmark of neurodegenerative diseases such as Alzheimer's and Parkinson's is the aberrant aggregation of proteins into amyloid fibrils, and small molecules that selectively bind to these fibrils hold promise as diagnostics, imaging probes, and therapeutics. Predicting how such ligands bind to fibril targets, however, presents two fundamental challenges. First, resolved co-crystal structures of amyloid-ligand complexes are exceptionally scarce; even with recent advances in cryo-EM only a handful have been structurally characterized, making supervised training of docking models impractical for this target class. Second, amyloid fibrils present a binding mode fundamentally different from globular proteins: ligands intercalate into longitudinal cross-$\beta$ grooves and stack cooperatively along the fibril axis, a geometry that existing docking models are not designed to capture. To address these challenges, we present CORAL (COopeRative Amyloid Ligand docking), a reinforcement learning framework that trains a generative docking model to produce ligand pose distributions tailored to the cross-$\beta$ groove geometry. Our reward explicitly incorporates cooperative ligand-ligand stacking energy alongside protein-ligand docking affinity, directly capturing the distinctive binding geometry of amyloid fibrils. We further introduce a curated evaluation set of amyloid-ligand complexes constructed from model-generated poses validated by domain experts. Experiments on both experimentally resolved structures and this evaluation set demonstrate improved pose quality and binding affinity correlation over existing docking baselines.

AI 总结：研究乘性噪声模型下任意范数压缩映射的随机逼近，提出统一基本分析方法，通过平均噪声序列等得到范数误差的李雅普诺夫漂移不等式，进而得出均方和集中性界，还讨论了证明技术的可推广性。

链接：https://arxiv.org/abs/2607.17595

机构：Department of Electrical Engineering, Stanford University（电气工程系，斯坦福大学）

作者：Siddharth Chandak

英文摘要： We establish mean-square and concentration bounds for stochastic approximation (SA) with arbitrary norm contractive mappings, under a multiplicative noise model where the noise may scale affinely with the norm of the iterates, and the iterates are potentially unbounded. These settings arise in reinforcement learning, where operators are often contractive in the $\ell_\infty$ norm and the noise scales with the iterates. To address the arbitrary norm, earlier works replace the non-smooth squared norm with a smooth Lyapunov function constructed via the generalized Moreau envelope. For concentration analysis, these works handle multiplicative noise and unbounded iterates through a multi-stage bootstrapping argument that starts from a time-varying worst-case bound and iteratively refines it. We instead present a unified and elementary analysis that yields both bounds. Using an averaged noise sequence and corresponding auxiliary iterates, we obtain a one-step Lyapunov drift inequality for the normed error directly, without smoothing the norm or constructing an envelope. For the mean-square bound, we combine this drift inequality with an induction argument showing that the iterates remain bounded in expectation. For the concentration bound, we develop a probabilistic induction over a sequence of "good" events on which the iterates are controlled, allowing the standard Azuma-Hoeffding bound to be applied. Our approach yields the first sub-Gaussian tailed maximal (all-time) concentration bound for SA under multiplicative noise, by allowing the stepsize to depend logarithmically on the confidence level. Beyond the specific setting considered here, we discuss the generalizability of these proof techniques to other noise models and iterative algorithms.

AI 总结：研究少样本逆强化学习（FM - IRL）问题，提出多任务判别器近邻引导的IRL（MPG）方法，通过学习两个互补奖励组件，在多种有显著变化的任务上验证有效性，平均成功率达81.2%，优于基线。

链接：https://arxiv.org/abs/2607.17760

作者：Ziyi Liu, Grace Zhang

英文摘要：Inverse reinforcement learning (IRL) provides a powerful framework for learning from demonstrations. However, real-world tasks often exhibit substantial natural variations (e.g., picking up mugs with varying shapes), making it impractical to collect demonstrations that fully specify a new task under every possible scenario. In practice, while demonstrations for the target task are limited, it is often easier to obtain datasets of heterogeneous but related behaviors. This motivates the problem of few-shot IRL with multi-task demonstrations (FM-IRL), where an agent must learn a new task with substantial variations from only a limited number of target-task demonstrations, together with sufficient demonstrations of related tasks and online agent experience. To do so, we must both recover the expert distribution of the new task and provide guidance when the agent deviates from it. We introduce Multitask discriminator Proximity-Guided IRL (MPG), which learns two complementary reward components: (1) a generalizable discriminator that transfers shared structure across related tasks to identify expert behavior in a new task, and (2) a proximity function that measures how far a state deviates from expert behavior and provides corrective guidance during exploration. We demonstrate the effectiveness of our method on multiple challenging navigation and manipulation tasks under significant variations (e.g., object configurations, table layouts, and initial robot poses), achieving an average success rate of 81.2%, outperforming the strongest per-task baseline by an average of 24.7 percentage points.

AI 总结：研究有限 horizon 强化学习中$\max$@$k$学习问题，指出其与标准预期回报最大化不同，证明马尔可夫策略不足，识别状态增强，表征策略性能差距，表明学习更难，给出高效算法实现最优样本复杂度率。

链接：https://arxiv.org/abs/2607.17823

机构：Bocconi University（博科尼大学）

作者：Riccardo Poiani, Martino Bernasconi, Andrea Celli

英文摘要：Reinforcement Learning is a cornerstone technique for modern large reasoning models. Usually, for difficult tasks such as code generation and theorem proving, the agent is evaluated by generating $K$ responses rather than sampling a single response, and performance is then measured using a retry-aware metric such as $\max$@$k$. Despite their practical importance, the theoretical foundations of learning under such criteria remain limited. In this work, we provide a theoretical study of the $\max$@$k$ learning problem in finite-horizon reinforcement learning. We show that optimizing the $\max$@$k$ objectives is fundamentally different from standard expected-return maximization. In particular, we prove that Markovian policies are in general insufficient, identify a compact state augmentation that restores optimality, and explicitly characterize the performance gap that can arise between history-dependent and non-history-dependent policies. Moreover, we show that learning $\max$@$k$-optimal policies is statistically harder than standard reinforcement learning and provide an efficient algorithm that achieves the optimal sample complexity rate.

AI 总结：研究分布软策略迭代中策略评估步骤里的固定策略分布软贝尔曼算子，基于克拉默几何，制定CDF级算子并证明其收缩性质，获得唯一不动点，还通过共轭得到谱域表示，为DSPI式算法研究提供参考。

链接：https://arxiv.org/abs/2607.17897

机构：School of Electrical and Electronic Engineering, University College Dublin（都柏林大学学院电气与电子工程学院）； School of Vehicle and Mobility, Tsinghua University（清华大学车辆与运载学院）； College of Artificial Intelligence, Tsinghua University（清华大学人工智能学院）

作者：Keru Wang, Yixin Deng, Yao Lyu, Stephen Redmond, Shengbo Eben Li

英文摘要：Distributional soft policy iteration (DSPI) provides an important framework for combining distributional reinforcement learning (DRL) with maximum-entropy control, in which the policy evaluation step is governed by a distributional soft Bellman operator acting on entropy-regularised returns. Theoretical analysis of such an evaluation step requires a probability metric under which Bellman updates can be controlled, typically by showing that the operator contracts the distance between any two candidate return-distribution estimates. In this paper, we focus on the Cramér geometry, a cumulative distribution function (CDF)-based metric with an $L^2$ structure, and study whether the fixed-policy distributional soft Bellman operator has this contraction property and hence a unique fixed point under this metric. Working directly on an admissible CDF field domain, we formulate the CDF-level distributional soft Bellman operator, prove that it is a $\sqrt{\gamma}$-contraction, and obtain the corresponding unique fixed point together with convergent iterative policy evaluation. The CDF formulation also shows that this finite-Cramér-domain property follows from a uniform first-moment condition on the combined one-step reward entropy shift, rather than from separate uniform boundedness assumptions on the reward and entropy terms. We then transport the same evaluation problem to the spectral domain by conjugation, obtaining an equivalent Hilbert-space representation of the same decision process. Taken together, these results identify the Cramér-geometric Bellman fixed point associated with the policy-evaluation step of DSPI, providing a reference point for studying approximate critics, evaluation error, and critic-loss design in DSPI-style algorithms.

AI 总结：研究大语言模型在RLHF训练中应对价值冲突的问题，核心方法是利用思维链推理，通过几何视角分析其作用，创建新的以价值冲突为重点的CoT设计，提升了道德推理性能，为改进模型处理复杂价值冲突请求的性能提供新途径。

链接：https://arxiv.org/abs/2607.17946

机构：University of Illinois - Urbana-Champaign（伊利诺伊大学厄巴纳 - 香槟分校）； Carnegie Mellon University（卡内基梅隆大学）

作者：Saket Reddy, Andy Liu

英文摘要：Large Language Models (LLMs) often struggle to navigate value conflicts when trained with the compressed scalar rewards of Reinforcement Learning from Human Feedback (RLHF). To address this challenge, we investigate how chain-of-thought (CoT) reasoning can help improve performance in this domain. Geometrically, we show that CoT correlates with further smoothing the model's loss landscape in its sharpest direction, helping resolve the optimization instability of traditional scalar rewards. We also demonstrate via relevant downstream benchmarks that value conflict-focused CoT may generalize to different kinds of moral reasoning, demonstrating that this CoT has the potential to be an effective mechanism for better moral reasoning. To capitalize on this potential, we create a new value conflict-focused CoT design that further smooths the sharpest direction of the loss landscape and increases moral reasoning performance. This finding shows that explicitly modifying and improving the design of reasoning dynamics offers a promising avenue for improving model performance on user requests with complex value conflicts, advancing pluralistic alignment in LLMs.

AI 总结：研究如何在强化学习中进行信息探索，提出基于贝叶斯核方法理论的随机特征信息增益（RFIG），用随机傅里叶特征近似信息增益，给出误差界并避免黑箱问题，实践细节使其可扩展到深度RL场景，实验显示其性能有竞争力且理论解释优。

链接：https://arxiv.org/abs/2607.17981

作者：Waris Radji, Odalric-Ambrym Maillard

英文摘要：Representation learning has enabled classical exploration strategies to be extended to deep Reinforcement Learning (RL), but often makes algorithms more complex and theoretical guarantees harder to establish. We introduce Random Feature Information Gain (RFIG), grounded in Bayesian kernel methods theory, which uses random Fourier features to approximate information gain and compute exploration bonuses in non-countable spaces. We provide error bounds on information gain approximation and avoid the black-box aspects of neural network-based uncertainty estimation, for optimism-based exploration. We present practical details that make RFIG scalable to deep RL scenarios, enabling smooth integration into standard deep RL algorithms. Experimental evaluation across diverse control and navigation tasks demonstrates that RFIG achieves competitive performance with well-established deep exploration methods while offering superior theoretical interpretation.

AI 总结：研究针对紧凑模型训练成本高问题，提出MADA-RL框架，将其分为生成器和评论家角色，用辩论感知信号训练，通过LoRA微调少量参数。核心是反事实评论家优势，提高了模型准确率，在数学推理基准测试中有显著效果。

链接：https://arxiv.org/abs/2607.18006

作者：Martino M. L. Pulici, Cuong Xuan Chu, Evgeny Kharlamov, Zifeng Ding, Volker Tresp, Yunpu Ma

英文摘要：Large language models achieve strong reasoning performance, but often at prohibitive training cost - a challenge that is especially acute for compact models ($\leq 4 \, \mathrm{B}$ parameters) trained under limited budgets. We introduce MADA-RL, a post-training framework that specializes compact models into generator and critic roles and trains them with a debate-aware learning signal, fine-tuning only a small subset of parameters via LoRA adapters. Our central contribution is a counterfactual critic advantage: a dynamic, role-conditioned baseline that redefines the critic's advantage as its reward minus the generator ensemble's per-instance accuracy. This explicitly optimizes critics to improve over generator consensus rather than to merely reproduce a correct answer, yielding more targeted credit assignment than static mean-reward normalization. At deployment, the specialized agents are composed in a lightweight multi-round protocol. Across five mathematical reasoning benchmarks, MADA-RL raises the accuracy of the DeepSeek-R1-Distill-Qwen-1.5B model from $39.9 \, \%$ to $41.9 \, \%$ ($+2.0$ points, $p < 0.001$) using $16$ times fewer trainable parameters than fully fine-tuned baselines, placing it on the accuracy-trainable-parameter Pareto front. It approaches, but does not surpass, the strongest baselines (DeepScaleR, STILL-3), which are trained on substantially larger datasets; we analyse this gap and the associated inference-time cost directly. A controlled study isolates the source of MADA-RL's gains: the counterfactual advantage produces the highest critic improvement rate of any model evaluated, indicating that trained critics learn to correct generator errors rather than to imitate them.

AI 总结：研究贝尔曼方程形式的来源，表明最优值函数递归性质源于三个条件，这些条件产生三种对偶性，框架揭示对偶性源于单一构造，统一了强化学习、控制和决策理论中相关方法。

链接：https://arxiv.org/abs/2607.18077

作者：Fernando E. Rosas, David Hyland, Daniel Polani

英文摘要：What gives the Bellman equation its form? We show that the recursive properties of optimal value functions follow from three conditions: that the dynamics decomposes through sufficient statistics, that the return decomposes recursively, and that the aggregation of uncertainty is compatible with both. When all three conditions hold on a common state, the Bellman equation arises from their mutual consistency; when one fails, tractability can often be recovered by augmenting the state or by deforming return or dynamics. The same conditions are shown to give rise to three dualities: one between probability and return, one between return and aggregation, and one between aggregation and probability. Our framework reveals these dualities as arising from a single construction, unifying methods developed separately across reinforcement learning, control, and decision theory.

AI 总结：研究基于评分标准的强化学习中探索有限问题，提出标准蒸馏策略优化（CriPO），通过策略内自蒸馏同时解决未探索标准和被抑制标准问题，在医学和科学基准测试中表现出色，减少优化步骤并提升性能。

链接：https://arxiv.org/abs/2607.18082

机构：Zhejiang University（浙江大学）； ByteDance（字节跳动）

作者：Mingxuan Xia, Yuhang Yang, Chao Ye, Shuai Zhu, Shenzhi Yang, Guangcheng Zhu, Yuhang Zhang, Cheng Peng, Haobo Wang, Siqing Wang

英文摘要：Rubric-based RL has recently shown promise in improving LLMs on open-ended tasks. A widely recognized limitation of rubric-based RL is limited exploration: criteria that no rollout manages to satisfy (Unexplored Criteria, UC) receive no optimization signal. Recent methods address this by incorporating rubric information as external guidance during rollout, yet they introduce a train-inference mismatch: the policy is optimized on rollouts produced under external guidance while this guidance is absent at inference time, causing error accumulation through autoregressive decoding. Moreover, these exploration-focused approaches overlook a fundamentally different failure mode that we term Suppressed Criteria (SC) -- criteria that are satisfied by some rollouts yet whose learning signals are lost during optimization because scalar reward aggregation assigns them non-positive aggregate advantages. Our analysis reveals that SC are remarkably prevalent: over 57% of samples exhibit this failure mode throughout training, with an average of 1.8 SC per sample. To simultaneously address both UC and SC without introducing training-inference mismatch, we propose Criterion-Distilled Policy Optimization (CriPO), which enhances rubric-based RL via on-policy self-distillation. For UC, CriPO constructs a criterion-injection self-teacher and computes a localized forward-KL loss to inject missing behaviors into the policy. For SC, CriPO employs a counterfactual self-teacher to locate criterion-relevant tokens in negative-advantage rollouts and flips their token-level advantages to positive values, preserving useful patterns that would otherwise be suppressed. Experiments on medicine and science benchmarks demonstrate that CriPO consistently outperforms rubric-based RL, achieving stronger final performance with approximately $2\times$ fewer optimization steps.

AI 总结：研究不可验证任务，提出体验式学习（EL），将LLM反馈模型从评判转为教练，通过提炼体验知识提供密集监督，在开放式任务上表现优于基于规则的RL，泛化性好且减轻奖励作弊。

链接：https://arxiv.org/abs/2607.18110

机构：Microsoft Research（微软研究院）； Tsinghua University（清华大学）； Peking University（北京大学）

作者：Tianzhu Ye, Li Dong, Guanheng Chen, He Zhu, Xun Wu, Shaohan Huang, Furu Wei

英文摘要：Reinforcement learning (RL) on open-ended tasks compresses an LLM's rubric-based evaluation into a scalar reward, discarding rich textual feedback and conflating responses with distinct quality profiles. We propose Experiential Learning (EL), which repurposes the feedback model from an LLM-as-a-Judge into an LLM-as-a-Coach. The coach distills its assessment of each on-policy response into transferable experiential knowledge, which conditions a teacher model and is internalized by the policy through on-policy context distillation. Compared with scalar rewards, this higher-bandwidth feedback channel provides dense supervision and preserves fine-grained preferences among high-quality responses. Across two policy families, with feedback from the policy itself or a proprietary model, EL consistently outperforms rubric-based RL on held-out and unseen open-ended tasks. Notably, EL generalizes better beyond the training distribution, and mitigates reward hacking. These findings establish experiential knowledge as a richer and more generalizable learning signal for post-training on non-verifiable tasks.

AI 总结：研究针对脑网络等随时间变化的复杂系统，提出TVGL-CFM模型，通过对数欧几里得图和条件流匹配，可生成和预测时变网络轨迹，在多数据集上表现优于原始信号基线，直接生成结构化精度轨迹更可靠。

链接：https://arxiv.org/abs/2607.16894

机构：University of Strathclyde（斯特拉斯克莱德大学）

作者：Om Roy, Yashar Moshfeghi, Keith Malcolm Smith

英文摘要：Many complex systems such as brain networks, financial markets, and gene-regulatory circuits are described not by a fixed graph but by one that changes over time. A standard way to summarise such structure at each instant is the sparse precision (inverse-covariance) matrix, and the time-varying graphical lasso (TVGL) turns a multivariate signal into a smooth chain of these matrices. We introduce TVGL-CFM, a single model that learns the distribution of such chains and can both generate new, realistic time-varying network trajectories for a given class and forecast how an observed trajectory will continue. Each precision matrix lives on a curved space of positive-definite matrices, but a log-Euclidean chart flattens an entire trajectory into an ordinary vector space, so a simple conditional flow-matching model can be trained and sampled there while every decoded matrix is guaranteed to be a valid precision matrix. For forecasting we start the flow not from noise but from a rough extrapolation of the recent history, so the model only has to learn a small correction. Across EEG motor-imagery, chaotic systems, and gene-expression data, TVGL-CFM generates trajectories that keep the class-discriminative structure of real data, and it forecasts future connectivity more accurately than raw-signal baselines. Generating the structured precision trajectory directly is therefore more faithful than generating raw signals and estimating connectivity afterwards.

AI 总结：针对高分辨率大气数据受限问题，Apeliotes框架基于全球再分析等数据构建，能提供千米尺度天气变量和多层大气场，经评估性能极具竞争力，预测垂直风廓线等有高精度表现。

链接：https://arxiv.org/abs/2607.17037

机构：Old Dominion University（奥多明尼昂大学）

作者：Evangelia Rafaela Frastali, Achyut Paudel, Maryam Golbazi, Frank Liu

英文摘要： High-resolution atmospheric data are required to resolve mesoscale and localized meteorological structures, however such datasets remain limited in many regions of the world. Existing high-resolution weather products are typically produced through dynamical downscaling, which is computationally expensive and difficult to scale across locations, variables, and forecast scenarios. These limitations motivate machine-learning-based downscaling systems that can generate multiple weather variables stochastically while producing new high-resolution fields directly. In this paper we present Apeliotes, a framework for high-resolution weather forecasting. Built on the global re-analysis atmospheric data, a pre-trained global weather foundation model, and a regionally trained generative diffusion model, Apeliotes not only provides accurate kilometer-scale weather variables, but also multi-level atmospheric fields which are not directly available in the existing global atmospheric data. Our comprehensive evaluation demonstrates that Apeliotes achieves highly competitive performance. The model predicts vertical wind profile with less than 3\% error between truth and predicted fields, achieving correlations of 0.91 for 10-m wind speed and 0.99 for 2-m temperature, with NRMSE values of 0.42 and 0.17, respectively.

AI 总结：针对不平衡回归难题，提出DADIR框架，含密度感知自适应划分、密度正则化条件变分自编码器和潜在空间数据平衡三个组件，能有效识别少数区域、保留稀疏区域信息并生成合成数据，提升不平衡回归数据集预测性能。

链接：https://arxiv.org/abs/2607.17178

机构：University of Zanjan（赞詹大学）

作者：Shermin Shahbazi, Hossein Mohammadi, Mohsen Afsharchi

英文摘要：Imbalanced learning addresses predictive modeling problems with underrepresented regions of the data distribution. Although widely studied in classification, imbalanced regression remains challenging because of continuous target variables and heterogeneous density distributions. Existing data-level methods often rely on fixed target partitioning or synthetic sample generation without jointly considering density variations and local feature-space structure. We propose DADIR, a Density-Aware Data-level Imbalanced Regression framework that exploits density information throughout the balancing process. DADIR comprises three components: (1) Density-Aware Adaptive Partitioning (DAAP), which recursively partitions the target space according to density variations; (2) a Density-Regularized Conditional Variational Autoencoder (DR-CVAE), which preserves sparse-region representations while learning latent features; and (3) latent-space data balancing, which combines feature-level clustering with oversampling to generate structurally consistent synthetic samples. Together, these components identify minority regions more effectively, preserve sparse-region information, and generate realistic synthetic data. The resulting balanced dataset can be used directly with existing regression models without modifying their architecture or learning objective. Experiments on diverse imbalanced regression datasets demonstrate consistent improvements in predictive performance, particularly in underrepresented regions, while also improving overall accuracy.

AI 总结：研究GRPO扩展到扩散模型的计算瓶颈问题，提出JAGG方法，通过特定插值和聚合梯度，减少反向传播次数，实验表明该方法能在质量损失小的情况下显著加速T2I训练中的DiT RL训练。

链接：https://arxiv.org/abs/2607.17572

作者：Ruiyi Ding, Jie Li, He Kang, Ziyan Liu, Chengru Song, Yuan chen

英文摘要：Group Relative Policy Optimization (GRPO) is a powerful reinforcement learning algorithm for aligning generative models with human preferences. While successful in large language models~\cite{shao2024deepseekmathpushinglimitsmathematical}, its extension to diffusion and flow matching models introduces a severe computational bottleneck: gradients must be back-propagated through the high-capacity DiT backbone at \emph{every} timestep of the sampling trajectory, making high-resolution text-to-image (T2I) training prohibitively expensive. Training-free DiT inference acceleration methods (e.g., $\Delta$-DiT, ScalingCache) exploit the fact that DiT hidden states and velocity predictions vary \emph{smoothly and nearly linearly} along the trajectory. We ask whether the same linearity can reduce the backward-pass cost of DiT RL training, and answer affirmatively with \textbf{JAGG} (\textbf{J}acobian-\textbf{A}ggregated \textbf{G}roup \textbf{G}radient), which reduces full transformer backward passes from $W$ to $2$ per group of $W$ consecutive steps. JAGG approximates intermediate-step Jacobians via $t$-weighted interpolation of the endpoint Jacobians, then aggregates per-step upstream signals into two composite gradients applied through a single joint backward pass. We prove this interpolation is \emph{exact} when the velocity is linear in $(z,t)$, and a cosine-similarity routing rule (\texttt{jagg\_frac}) deploys JAGG only where the assumption holds. Experiments on T2I benchmarks show JAGG delivers $\sim$2$\times$ backward speedup with negligible quality degradation.

AI 总结：研究提出DiFA框架，将扩散模型推理时的数据预测细化转为顺序状态估计问题，受卡尔曼滤波启发建立前向对齐时间一致性，引入偏差引导机制，实验表明该方法在多个指标上显著提升CIFAR-10和ImageNet的生成保真度。

链接：https://arxiv.org/abs/2607.17972

作者：Shigui Li, Delu Zeng

英文摘要：The prevailing inference framework for diffusion models formulates generation fundamentally as a problem of numerical integration. This perspective casts the model as an exact estimator, neglecting the inherent statistical uncertainty of the denoising process. In this work, we propose Forward-Process Aligned Diffusion prediction (\textbf{DiFA}), a training-free framework that reframes inference-time data prediction refinement as a sequential state estimation problem. Rather than reusing past outputs solely for numerical integration, DiFA treats iterative data predictions along the reverse trajectory as correlated observations to build a forward-aligned temporal consensus. Inspired by Kalman filtering, this consensus aggregates historical predictions according to structural consistency and noise-level compatibility. To counteract the over-smoothing tendency of temporal consensus, we introduce a deviation guidance mechanism to adaptively preserve residual details. Empirically, DiFA yields significant improvements on CIFAR-10 and ImageNet across the evaluated metrics, including FID, IS, and FD-DINOv2, demonstrating that aligning inference with the forward statistical structure substantially improves generative fidelity.

AI 总结：研究探讨通用语言模型在3D分子设计中应对复杂空间约束的能力，引入3D - Fit评估策略，发现语言模型虽落后于先进方法，但有潜力同时处理多种空间约束。

链接：https://arxiv.org/abs/2607.18144

作者：Thomas MacDougall, Maksim Kuznetsov, Roman Schutski, Rim Shayakhmetov, Maxim Malkov, Vladimir Aladinskiy, Alex Aliper, Alex Zhavoronkov

英文摘要：Structure-based drug design (SBDD) leverages the 3D structure of protein targets, often complemented by other spatial constraints, to generate candidate binding molecules. While diffusion models have dominated as a leading paradigm for high-quality 3D molecule generation, LLM-based methods are rapidly emerging in molecular design and have shown competitive performance in pocket-conditioned molecular generation. However, their ability to reason about physics and 3D spatial environments is largely underexplored. In this work, we systematically analyze whether current general-purpose LLMs are capable of navigating complex 3D constraints compared to established baselines such as specialized diffusion models. We consider 3D ligand generation conditioned on protein pockets together with ligand- and interaction-derived spatial constraints, including anchor fragments, pharmacophore points, and mandatory pocket-ligand interactions. To enable this evaluation, we introduce 3D-Fit - a token-efficient benchmarking strategy for assessing LLM performance on multi-conditioned spatial molecule generation. Our findings reveal a clear pattern in LLM spatial capabilities: while they still lag behind state-of-the-art approaches, they are promising and can handle multiple spatial constraints simultaneously, enabling scaling to heterogeneous setups.

AI 总结：研究提出三体散射建模（TBSM）用于生成，将能量距离转化为特定相互作用，为单步生成器提供监督。通过在线跟踪条件期望减少噪声，在ImageNet-256上训练单步生成器取得良好结果，确立了跟踪散射作为高维单步生成的途径。

链接：https://arxiv.org/abs/2607.18198

机构：Westlake University（西湖大学）； Zhejiang University（浙江大学）； University College London（伦敦大学学院）

作者：Peng Sun, Zhenglin Cheng, Deyuan Liu, Jun Xie, Xinyi Shang, Tao Lin

英文摘要：Modern generative models typically rely on an adversarial critic, a prescribed noise-to-data path, or an autoregressive factorization. Instead, we show that a proper distributional energy can induce sample-level motion and provide direct regression supervision for a one-step generator. Three-Body Scattering Modeling (TBSM) for generation turns the energy distance into a constant-size per-projectile interaction: each projectile is attracted toward one real source and repelled from one independently generated source. Conditioned on the projectile and its condition, its expectation equals the $2$-Wasserstein gradient-flow velocity of $\frac12D_E^2(P_{\theta},Q)$. A batch of $B$ frozen-target events yields $O(B)$ sample-level losses, each using one reference for its condition instead of the minibatch-wide all-pairs field used by methods such as Drifting Models. Tracking this conditional expectation online can reduce field noise. Using scattering in frozen image features, TBSM trains one-step generators on ImageNet-256, achieving FID${}=2.23$ with pixel-space PixelDiT-XL and FID${}=1.63$ with latent-space DiT-XL at NFE${}=1$. We provide a design map relating diffusion-related supervision, Drift-like dynamics, and GAN-like objectives. These results establish tracked scattering as a route to high-dimensional one-step generation. Code: this https URL.

AI 总结：研究旨在解决不规则时间序列的因果发现问题，通过扩展PCMCI+方法，利用预定义时间窗口聚合因果影响，在合成数据实验中表现出色，能有效恢复因果图且优于标准方法。

链接：https://arxiv.org/abs/2607.18226

机构：Feedzai（Feedzai公司）； Instituto Superior Técnico, Universidade de Lisboa（里斯本大学高等技术学院）； DCC, Faculdade de Ciêncidas da Universidade do Porto（波尔图大学理学院数据与计算机科学系）； Instituto de Telecomunicações（电信研究所）

作者：Martim Penim, Ricardo Ribeiro Pereira, Jacopo Bono, Hugo Ferreira, Mário A.T. Figueiredo, Pedro Bizarro

英文摘要：Causal discovery methods have shown strong performance in temporal systems, but they typically rely on regular and discrete lag structures, limiting their applicability to regularly sampled data. However, many real-world tasks require dealing with irregularly sampled streams of events, such as sensor streams, healthcare data, and financial transactions. In this work, we propose an extension of PCMCI+, a state-of-the-art method for causal discovery on regular multivariate time series, to allow for handling irregular time series. Instead of modelling causal relations through fixed-lag dependencies, our method aggregates causal influence over predefined temporal windows. We evaluate our method on synthetic irregular event streams with known causal structures under different signal-to-noise ratios, showing that it consistently recovers the underlying causal graph and substantially outperforms the standard PCMCI+ on irregularly sampled data.

AI 总结：研究随机优化中样本级危害问题，提出将参数更新形式化为优化问题，引入高效代理，通过降维、限制优化层等方法减少内存和速度瓶颈，集成到标准优化器，实验证明能减少样本干扰、提升泛化能力。

链接：https://arxiv.org/abs/2607.16261

机构：Amadeus（阿马德乌斯）

作者：Apostolos Avranas

英文摘要： Modern optimizers combine gradients from the current mini-batch with historical optimization state, such as momentum or adaptive moments. While highly effective, aggregating across the batch and incorporating this history can produce parameter updates that increase the loss of individual samples. We term this effect harm and formalize the parameter update as an optimization problem that explicitly minimizes the conflicting impact of both batch averaging and past optimization state on current data. Because the exact formulation is intractable, we introduce a highly efficient proxy. We first reduce the problem's dimensionality to the batch size, and then drastically cut memory and speed bottlenecks by successfully restricting the optimization to the last linear layer. This hinges on the unexpected finding that this layer alone reliably captures the second-order statistics of the per-sample gradients. The resulting surrogate problem integrates readily into standard optimizers like SGD and AdamW, and can be solved using a small number of GPU-friendly iterations. Crucially, the method exhibits favorable scaling properties, as the relative computational overhead shrinks as the model size or input grows. Experiments on image classification benchmarks confirm reduced per-sample interference and improved generalization.

AI 总结：研究常步长随机梯度下降在平坦极小值处的缩放极限，通过分析由压缩驱动链生成马尔可夫噪声的SGD，证明其在特定距离下的收敛性，给出不同平坦指数下的收缩因子及小步长缩放极限，揭示了与强凸情况不同的行为。

链接：https://arxiv.org/abs/2607.16384

机构：Georgia Institute of Technology（佐治亚理工学院）

作者：Jingyi Zhang, Cheng Mao, Debankur Mukherjee

英文摘要：For stochastic gradient descent (SGD) with a constant stepsize $\alpha$, the invariant law of the iterates, centered at a minimizer, describes the behavior of the algorithm over long time horizons. In the strongly convex case, this invariant law has the familiar $\sqrt{\alpha}$ scaling and a Gaussian limit as $\alpha\downarrow 0$. We show that this behavior changes fundamentally for convex objectives $H$ with flat minima and (sub)quadratic tails. More specifically, we study SGD with Markovian noise generated by a contractive driving chain. For every sufficiently small constant stepsize $\alpha$, we prove existence, uniqueness, and geometric convergence to an augmented invariant law in a Wasserstein distance induced by an $\alpha$-dependent metric. When the minimizer $x_\star$ has local flatness exponent $m\ge2$, meaning that $\nabla^2 H(x)\asymp \lVert x-x_\star\rVert^{m-2} I_d$ as $x\to x_\star$, we obtain a contraction bound with factor $1-c\alpha^{m-1}$, where $c>0$ is a constant. This recovers the factor $1-c\alpha$ in the quadratic case $m=2$. We then analyze the small-stepsize scaling limit. We show that the invariant law concentrates on the scale $\alpha^{1/m}$ and that the rescaled iterates converge weakly to the stationary distribution of the stochastic differential equation $$ dY_t=-h_0(Y_t)\,dt+\Sigma^{1/2}\,dB_t, $$ where $h_0$ is the limiting drift at the minimizer and $\Sigma$ denotes the asymptotic covariance. This recovers the Gaussian limit when $m=2$ and gives generally non-Gaussian stationary limits in the flat case $m>2$. Finally, we give corresponding results for coordinate-separable objectives with unequal flatness exponents.

AI 总结：研究两层ReLU网络中宽度依赖模型超参数和$\ell_2$正则化对损失景观的影响，推导超参数条件及解析解，通过实验发现AdamW可防参数坍缩，揭示了相关因素对损失景观几何结构的影响。

链接：https://arxiv.org/abs/2607.16720

机构：Okinawa Institute of Science and Technology（冲绳科学技术研究院）

作者：Haruka Eshima, Makoto Yamada

英文摘要：Understanding deep neural networks remains a central challenge in machine learning. In particular, the theoretical properties of even two-layer ReLU networks, especially in the presence of weight decay, remain poorly understood. To this end, we derive a sufficient condition on the hyperparameter settings under which the global minima collapse to the zero solution. Interestingly, our experiments reveal that using AdamW as an optimizer prevents the collapse of the learned parameters, whereas using SGD does not, which may help explain the success of AdamW in deep learning training. In addition, when restricting the input dimension to one, we derive an analytical solution for the globally optimal parameter sets of two-layer ReLU networks and show that $\ell_2$-regularization has a width-invariant effect on connectivity, but its dimensionality-reducing effect becomes stronger as the network width increases. These results provide insight into how width-dependent hyperparameters influence the geometry of regularized loss landscapes.

AI 总结：研究利用机器学习方法结合独特特征选择算法预测多囊卵巢综合征，介绍结合特征工程和机器学习的数据驱动方法，经多种特征选择方法训练模型，结果显示经随机森林特征重要性和最高相关性选择十个特征的AdaBoost测试准确率最高。

链接：https://arxiv.org/abs/2607.16941

作者：Al Zadid Sultan Bin Habib, Md Asif Bin Syed, Md. Ekramul Islam, Tanpia Tasnim

英文摘要：Polycystic Ovarian Syndrome (PCOS) is a widespread hormone problem for women of childbearing age. Women with PCOS may not ovulate; they might have high levels of androgens and have many small cysts on the ovaries. It can cause missed or irregular menstrual periods, excess hair growth, acne, infertility, and weight gain. Machine Learning (ML) can effectively diagnose this disease at an earlier stage as tons of medical data are available now. Traditional approaches to detect PCOS encompass a combination of clinical evaluation, medical history assessment, physical examination, and laboratory tests. These approaches aim to identify the characteristic symptoms and hormonal imbalances associated with PCOS. Physical examination requires good resources and costs time and money. In recent times, data-driven techniques have substantially advanced disease prediction within the medical field. We aim to utilize ML approaches, incorporating unique feature selection algorithms, to predict PCOS. This paper introduces a data-driven approach to PCOS diagnosis, combining Feature Engineering and ML. Several feature selection approaches have been considered to select sets of features for training the ML model, including CatBoost, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), AdaBoost, Random Forest (RF). Results demonstrate that AdaBoost, with ten features selected by RF Feature Importance and Highest Correlation (HC), provides the highest test accuracy.

AI 总结：针对预测性维护多元时间序列数据的生存分析，提出SurvCF(t)框架，通过识别资产运行历史变化增加预测寿命，将解释设为约束优化问题，经多基准评估，弥合生存预测与维护差距，实现可解释决策型AI。

链接：https://arxiv.org/abs/2607.16969

作者：Zara Karazian, Panagiotis Papapetrou, Sindri Magnússon, Erik Frisk, Tony Lindgren

英文摘要： Predictive maintenance relies on accurate Remaining Useful Life estimation, often formulated using survival analysis over multivariate time-series data. While modern deep survival models achieve strong predictive performance, their black-box nature limits their use in safety-critical settings where actionable insight is required. In this work, we introduce \textit{SurvCF(t)}, the first framework for generating counterfactual explanations for survival models operating on time-series data. \textit{SurvCF(t)} identifies minimal, plausible, and temporally consistent changes to an asset's operational history that increase its predicted life time, framing explanation as a constrained optimization problem combining validity, proximity, sparsity, and plausibility. We evaluate the method on multiple benchmarks, including C-MAPSS, N-CMAPSS, and a real-world case study of the Scania Component\_X dataset, demonstrating its ability to produce actionable and interpretable interventions. Our results show that \textit{SurvCF(t)} bridges the gap between survival prediction and prescriptive maintenance, enabling explainable and decision-oriented AI for maintenance strategies.

AI 总结：该研究针对非靶向 LC-MS 代谢组学预处理决策多且结果依赖未知选择的问题，提出多宇宙共识管道，通过十阶段质量控制及多方法组合分析，在示范数据集上得出更稳健的特征选择结果，还讨论了方法的范围与局限。

链接：https://arxiv.org/abs/2607.17345

机构：Istanbul Medipol University（伊斯坦布尔梅迪波尔大学）； Istinye University（伊斯坦布尔伊斯蒂尼耶大学）

作者：Mohammed Saeed Al-Huraibi, Ihsan Yozgat, Ahmet Kaplan

英文摘要：Background: Untargeted LC-MS metabolomics requires a long chain of preprocessing decisions, each with several equally defensible options. Analysts typically commit to one pipeline and report the resulting feature shortlist. How strongly that shortlist depends on choices that were never varied stays invisible. Results: We adapt multiverse analysis to untargeted metabolomics feature selection. We present an auditable, configuration-driven pipeline that (i) applies a ten-stage quality-control filter cascade in which every feature's fate is logged, and (ii) runs the downstream analysis as a multiverse over four contrasting preprocessing philosophies, each combined with four feature-ranking methods under bootstrap stability selection and label-permutation testing. Only features recurring across paths enter a tiered consensus. On a demonstration dataset of five breast-cancer cell lines (30,370 detected features), the four single pipelines individually returned shortlists of 4-20 features whose pairwise agreement was as low as Jaccard = 0.05. The multiverse consensus retained 15 features (>=2/4 paths), of which one recurred across all four, although two paths (sharing normalization and drift-correction methods) dominate the consensus. A pipeline-wide label-permutation test found no false discoveries in 50 null permutations. Conclusions: Reporting only preprocessing-robust features, with a complete kept/dropped audit trail, converts hidden analytical degrees of freedom into an explicit, inspectable output. We discuss scope and limitations, including single-batch design and the need for independent validation.

AI 总结：研究随机非凸优化能否黑盒简化为在线凸优化的静态遗憾最小化，提出维护梯度跟踪器并由在线学习器选预处理器的方法，对光滑和非光滑非凸目标均适用，解决了相关开放问题，还为理解自适应优化方法提供新视角。

链接：https://arxiv.org/abs/2607.17607

机构：Center for Computational Science and Engineering, MIT（麻省理工学院计算科学与工程中心）； Department of Civil and Environmental Engineering, MIT（麻省理工学院土木与环境工程系）； Institute for Data, Systems, and Society, MIT（麻省理工学院数据、系统与社会研究所）

作者：Haichen Hu, David Simchi-Levi

英文摘要：We study whether stochastic nonconvex optimization can be reduced to ordinary static regret minimization in online convex optimization in a black-box manner. For smooth nonconvex objectives, our reduction maintains a predictable gradient tracker, while a black-box online learner selects a preconditioner that determines how this tracker is transformed into the update direction. The learner receives linear convex losses and is evaluated against a single fixed comparator over one undiscounted online game. For a $\beta$-smooth objective with range bounded by $M$ and an unbiased stochastic-gradient oracle with variance bounded by \(\sigma^2\), we establish $$\frac{1}{T}\sum_{t=1}^T \mathbb E\!\left[\|\nabla f(x_t)\|_2^2\right] \lesssim \frac{\sigma\sqrt{M\beta}}{\sqrt T} + \frac{\sqrt{M\beta}\, \mathscr R_T(\mathcal A,I_d)}{T} + \frac{M\beta}{T}.$$ Consequently, any black-box OCO algorithm with $\mathscr R_T(\mathcal A,I_d)=O(\sqrt T)$ recovers the classical $O(\frac{1}{\sqrt{T}})$ convergence rate. We further show that the same black-box framework extends beyond the smooth setting to Lipschitz nonconvex objectives without Lipschitz continuous gradients. Importantly, this extension continues to rely only on an ordinary static-regret guarantee and requires no stronger notion of online regret. When the OCO oracle admits square-root static regret, the resulting conversion achieves the optimal $O(T^{-2/7})$ convergence rate for the corresponding Goldstein stationary point. These results resolve the open problem posed by Chen and Hazan (2024). More broadly, our framework separates optimizer design into gradient prediction and online preconditioner selection, providing a principled perspective on how adaptive optimization methods such as AdaGrad and Shampoo may be understood through static regret and applied in nonconvex optimization.

AI 总结：研究离策略蒸馏中训练数据、目标参数化和模型能力的相互作用，通过分解问题进行“目标到能力”和“数据到目标”分析，引入诊断指标量化张力，探讨自适应目标路由，表明有效路由取决于信号质量，将持续预训练重构为监督设计问题。

链接：https://arxiv.org/abs/2607.16246

机构：Baidu Inc.（百度公司）

作者：Jiangan Yuan, Zhixuan Li, Han Xu

英文摘要： Off-policy distillation is now central to large language model pre-training, yet how training data, objective parameterization, and model capabilities interact remains poorly characterized. We studies top-$k$-truncated, temperature-scaled off-policy distillation by decomposing this problem into two questions: an \emph{objective-to-capability} analysis of how the training objective shapes token-level supervision and downstream performance, and a \emph{data-to-objective} analysis of how data heterogeneity should inform objective routing. We first show that the language-modeling objective ($L_{\mathrm{LM}}$) and the knowledge-distillation objective ($L_{\mathrm{KD}}$) induce systematically different capability profiles, and trace this divergence to a gradient-level tension between \emph{direct observed-token reinforcement} and \emph{teacher-supported alternative supervision}. To quantify this tension, we introduce diagnostic metrics -- support coverage, observed-token probability mass, and teacher-distribution concentration -- and show via controlled sweeps that the support size $k$ governs a coverage-sharpness trade-off, while distillation temperature controls within-support probability allocation. We then examine adaptive objective routing: a domain-level policy that applies $L_{\mathrm{LM}}$ to math and code and $L_{\mathrm{KD}}$ to general-domain data yields consistent gains over both single-objective baselines, whereas token-level routing based on observed-token probability mass or teacher entropy fails to consistently match the single-objective baseline. These results suggest that effective objective routing depends less on routing granularity than on the quality of the routing signal, reframing continued pre-training via off-policy distillation as a structured, data-conditional supervision-design problem rather than a global hyperparameter choice.

AI 总结：研究针对在线策略知识蒸馏的问题，提出CADENCE统一框架，其DRIFT机制结合六个扩展组件，能解决冷启动崩溃等问题。在GSM8K和MATH-500上实验，将大教师模型蒸馏成小学生模型，大幅提升性能，且无需数据中心规模硬件。

链接：https://arxiv.org/abs/2607.16955

作者：Satyam Kumar, Saurabh Jha

英文摘要：On-policy knowledge distillation transfers reasoning from large teachers to compact students, but existing approaches suffer three compounding failure modes: (i) cold-start collapse, where a fresh student assigns near-zero mass to teacher-preferred tokens; (ii) state-agnostic divergence scheduling, where time-only forward/reverse-KL interpolation ignores the student's coverage state; and (iii) binary reward sparsity, where pass/fail signals discard information from partially correct traces. We present CADENCE, a unified framework with a targeted fix for each. Its DRIFT mechanism schedules a per-token convex mixture of forward-KL and reverse-KL surrogate objectives on student-sampled trajectories (per-token surrogates, not sequence-level KL gradient estimators). Six components extend it: (A) COVA, a coverage-adaptive $\beta$ schedule accelerating the forward-to-reverse transition; (B) FTB, a forking-token boost concentrating gradient at high-entropy positions via a globally-normalized entropy reference; (C) CCD, a dense reward adding numerical-proximity partial credit for incorrect-but-close traces; (D) LAP, brevity-preferential correct-rollout reinforcement; (E) EMR, an entropy-matching calibration regularizer; (F) BSD, a bootstrapped self-distillation phase. On GSM8K and MATH-500 (corrected 512-token protocol, 5 seeds, reported std), CADENCE distills a 0.5B student from a 1.5B teacher to 69.8 $\pm$ 0.5% GSM8K pass@1 (from 48.7% pretrained; 63.2% of the teacher gap closed) and to 72.1 $\pm$ 0.4% with a 3B teacher (76.2% closed), beating the strongest matched-compute label-using baseline (DRIFT+binary reward) by +4.4 $\pm$ 0.7 points. All experiments run on a single Apple Mac Studio (M-series, 64GB unified memory), showing principled distillation reaches strong reasoning quality without datacenter-scale hardware.

AI 总结：针对汉塔病毒基因组监测受数据分布等限制的问题，提出HantaWatch联邦学习框架，集成多种技术，通过实验验证其能支持多任务并平衡多种性能，还可转化模型输出，为分散监测提供实用决策支持层。

链接：https://arxiv.org/abs/2607.16234

机构：School of IT, Deakin University（迪肯大学信息技术学院）

作者：Shanika Iroshi Nanayakkara, Shiva Raj Pokhrel

英文摘要：Hantavirus genomic surveillance is limited by the distribution of sequence data, non-IID source heterogeneity, and constrained expert-review capacity. We propose HantaWatch, a federated learning framework that enables laboratories and surveillance sites to collaboratively train sequence-based models without sharing raw data. HantaWatch integrates k-mer feature extraction, source-aware federated client construction, adaptive DU-FedProx optimization, surveillance-specific model selection, and prediction-only triage. Experiments on binary and multi-class tasks show that HantaWatch supports high-risk screening, outbreak-associated prediction, clade classification, and clinical-syndrome categorization while balancing predictive performance, false-negative risk, and update stability. The framework converts model output into risk scores, confidence estimates, uncertainty flags, and ranked expert-review priorities. HantaWatch therefore provides a practical federated decision-support layer for decentralized Hantavirus surveillance, supporting expert prioritization without replacing laboratory or public-health interpretation.

AI 总结：研究差分隐私机器学习中群体公平性问题，提出基于隐私成本公平比率（PCER）的度量方法，该方法仅需每组训练和测试准确性，通过实验评估其与标准公平性度量的差异，揭示了基于结果度量遗漏的模式及隐私保证对审计的影响。

链接：https://arxiv.org/abs/2607.16620

作者：Rakshit Naidu

英文摘要： Differential privacy (DP) is increasingly deployed to limit membership inference risk in machine-learning systems. Prior work has shown that DP-SGD can widen accuracy disparities across demographic groups, but this framing treats fairness as a purely outcome-side concern. We argue that privacy cost, the information leakage borne by each group, is itself a form of harm, and adopt a compensatory-fairness framework in which a group that involuntarily bears greater privacy exposure is owed proportionally greater benefit from the system. From this principle we derive the \emph{Privacy-Cost Equity Ratio} (PCER), a group fairness metric defined as a group's positive prediction rate normalized by its per-group overfitting gap. By a standard membership inference bound, this overfitting gap upper-bounds each group's vulnerability to inference attacks, making PCER a conservative measure of benefit relative to exposure. PCER needs only per-group train and test accuracy (no shadow models), making it a practical post-hoc audit tool. We evaluate PCER alongside standard fairness metrics across six benchmark--attribute combinations spanning tabular and NLP domains, under DP-SGD at a range of privacy budgets, and validate the overfitting-gap proxy against a direct threshold membership-inference attack. The results reveal patterns that outcome-based metrics miss. On COMPAS, PCER uncovers a persistent double disadvantage: the protected group bears both greater privacy exposure and worse predictive outcomes, something demographic parity gap masks entirely. Sensitivity analysis shows very strong privacy guarantees collapse both groups' overfitting to a numerical floor, rendering exposure-based audits uninformative in that regime. Together, these findings show that fairness audits of privacy-preserving systems must account for who bears the cost of protection, not only who benefits from its outcomes.

AI 总结：研究针对2型糖尿病风险预测模型外部测试和公平性评估不足问题，开发多维框架，用NHANES数据训练XGBoost模型并在BRFSS上验证，发现存在性能损失和公平性偏差，确定主要风险驱动因素，强调需公平性感知、年龄分层的部署策略。

链接：https://arxiv.org/abs/2607.16253

作者：Rajveer Singh Pall, Sameer Yadav, Siddharth Bhalerao, Sourabh Sahu, Ritu Ahluwalia, Bhaskar Awadhiya

英文摘要：Machine learning-based Type 2 diabetes risk prediction models obtain good internal validation results but lose effectiveness in real-world applications due to deficient external testing and fairness assessment. We developed a multi-dimensional framework evaluating discrimination, calibration, interpretability, and algorithmic fairness on nationally representative populations. An XGBoost model was trained on NHANES 2015-2020 (n=15,685) using eight non-laboratory predictors: age, sex, race/ethnicity, BMI, smoking status, physical activity, history of heart attack, and history of stroke. External validation was performed on BRFSS 2020-2022 (n=1,285,783) under realistic distribution shift. Internal validation showed good discrimination (AUC=0.794, 95% CI 0.788-0.800), with performance loss on external validation (AUC=0.717, relative decrease: -9.7%, p<0.001). Fairness analysis revealed severe bias: elderly adults (>=60) showed AUC=0.607 vs 0.742 for young adults (difference=0.135, p<0.001); obese individuals showed AUC=0.698 vs 0.735 for normal weight (difference=0.037, p<0.001). Gender showed comparable performance (male=0.723 vs female=0.712, p=0.142). Calibration revealed risk overestimation (Brier score=0.123). SHAP analysis identified age, BMI, and physical activity as primary risk drivers. Populations with highest diabetes risk receive the worst algorithmic performance, underscoring the need for fairness-aware, age-stratified deployment strategies before clinical use.

AI 总结：研究利用结构化国家层面每周数据，通过可解释机器学习框架预测呼吸疾病率和空气质量状况，比较多种回归和分类模型，经SHAP值解释及亚组分析发现PM2.5是主要预测因子，表明仅模型准确性不足以进行气候 - 健康预测，可解释模型有重要作用。

链接：https://arxiv.org/abs/2607.17024

作者：Maede Azani Hassan Abadi, Shouyi Wang

英文摘要：Air pollution and climate-related stressors are increasingly important concerns for respiratory health, especially in settings with unequal environmental exposure and healthcare capacity. This study evaluates an interpretable machine learning framework for predicting respiratory disease rates and air-quality status using structured country-level weekly data. Two supervised learning tasks were considered: regression of respiratory disease rate per 100,000 population and binary classification of air-quality status. Nine regression models and nine classification models were compared using nested cross-validation. Model interpretation was conducted using SHAP values, and subgroup analysis was performed across income levels and geographic regions. The results showed that PM2.5 concentration was the dominant predictor of respiratory disease rate, with linear and regularized linear models achieving the strongest regression performance. For air-quality classification, models achieved high balanced accuracy when PM2.5 was included, but performance decreased substantially when PM2.5 was removed, indicating strong dependence on pollutant-related information. SHAP analysis showed that, without PM2.5, socioeconomic and meteorological variables such as GDP per capita, precipitation, and healthcare access became more influential. Subgroup analysis showed similar aggregate regression error across income groups, but PM2.5 contributed more strongly to predictions in lower-middle-income countries. These results show that model accuracy alone is not sufficient for climate-health prediction. Interpretable models can help identify dominant pollution-related signals, test whether results depend on key pollutant variables, and show whether prediction patterns differ across socioeconomic groups.

AI 总结：研究线性注意力在关联记忆上的退化问题，提出核化线性注意力激活框架KATA，通过自对偶齐次锥认证非负权重导出特征图，实现高效计算，在长距离记忆等任务中表现出色，揭示了特征相关权衡及相互作用。

链接：https://arxiv.org/abs/2607.17419

作者：Ayoub Ghriss, Sourav Chakraborty

英文摘要：Linear attention promises constant-time recurrent inference but degrades sharply on associative recall. We formulate attention recall as a spherical-packing problem and introduce Kernelized Linear Attention Activations (KATA), a framework whose feature maps are derived from first principles by certifying nonnegative attention weights through a self-dual homogeneous cone. Building on this observation, we show that rank-one positive semi-definite (PSD) features offer a favorable capacity--interference tradeoff. KATA recovers a parameter-free convex output gate and characterizes associative capacity through the Welch interference floor. For tolerances above this floor, KATA enlarges the state without adding parameters and admits spherical codes with exponentially many keys in the projection dimension. We implement KATA as fused Triton kernels at two operating points: a flash-attention-style forward up to ${\sim}1.6\times$ FlashAttention-2 throughput, and an exact $O(T)$ chunked-state form that reaches ${\sim}11\times$ FlashAttention-2 forward throughput at $131$k tokens. An associative scan of the first-order feature lowers the inter-chunk recurrence depth to $O(\log(T/C))$ for chunk size $C$ and averages ${\sim}2.4\times$ the throughput of a matched sequential linear-attention baseline. On long-range MQAR and repeated-key overwrite, several KATA variants outperform Gated DeltaNet, with parameter counts and state sizes reported alongside accuracy. Induction preserves near-perfect recall, while kernel benchmarks show that the maps can be implemented efficiently. KATA retains $0.985$ MQAR at a $16\times$ out-of-distribution length, approaching the softmax with roughly one quarter of the KV-cache entries. Experiments on 340M-parameter LLMs reveal a feature-dependent fluency trade-off and clarify how positional embeddings, delta rules, and decay gates interact with feature geometry.

AI 总结：研究旨在预测MCI到AD的转化，将持久同调用于临床轨迹点云，结合多种特征构建堆叠集成模型。校正泄漏源，经交叉验证等实验，模型有竞争力的准确性，提出H0持久熵为拓扑生物标志物，实现个体层面不确定性量化。

链接：https://arxiv.org/abs/2607.17442

机构：Institute of Health Informatics, University College London（伦敦大学学院健康信息学研究所）

作者：Navin Bondade

英文摘要：Background. Predicting conversion from mild cognitive impairment (MCI) to Alzheimer's disease (AD) is central to trial enrichment and care planning, yet existing models provide no individual-level uncertainty estimates and rarely include transparent leakage audits. We introduce the first application of persistent homology to longitudinal clinical trajectory point clouds for this task, and the first split-conformal individual risk guarantee for any AD-conversion model. Methods. We analysed 741 MCI subjects (240 converters, 32.4%) from ADNI with a uniform 4-year follow-up cap. Five leakage sources were corrected; without them a naive pipeline achieved AUC=0.934, inflated by +0.075. Vietoris-Rips persistent homology and sublevel-set proxies were combined with trajectory slopes and engineered features (76 total) in a stacking ensemble evaluated by 5-fold cross-validation. Results. Cox and Random Survival Forest models with TDA features achieved concordance C=0.799 and C=0.826 versus C=0.753 and C=0.812 without (+0.045 and +0.014). The primary nested AUC is 0.840 (same-fold bound 0.866); external AUC was 0.879 on a zero-overlap ADNI-2/GO/3 cohort. H0 persistence entropy was the top SHAP feature and significantly associated with APOE4 dosage (Spearman r=-0.191, p<0.0001, Bonferroni-corrected). Cross-conformal coverage was 90.4%+-2.2% (target 90%); empirical external coverage 96.9%. Maximum fairness gap in false-negative rate across seven subgroups was 0.092. Conclusions. We propose H0 persistence entropy as a topological biomarker of cognitive decline and demonstrate that a leakage-audited, conformally calibrated pipeline reaches competitive accuracy with individual-level uncertainty quantification not previously available for this task.

AI 总结：研究探讨变压器对给定任务是否最优，提出为数据集优化变压器架构的方法，在算法玩具任务和代码语言建模数据集上实验，发现标准变压器非局部最优，简单替代方案有优劣，暗示有改进架构可支持多种能力。

链接：https://arxiv.org/abs/2607.17624

机构：Idiap Research Institute（Idiap研究所）； EPFL（洛桑联邦理工学院）； Adelaide University（阿德莱德大学）

作者：Damien Teney, Liangze Jiang, Hemanth Saratchandran, Simon Lucey

英文摘要：Transformers are remarkably versatile and their design is largely consistent across a variety of applications. But are they optimal for any given task or dataset? The answer may be key for pushing AI beyond merely scaling current designs. *Method.* We present a method to optimize a transformer architecture for a given dataset, which we use as a tool to study optimal task-specific inductive biases. This method replaces the most important non-linearities (GeLUs,;softmax) with functions learned on held-out data. We then train the resulting architectures on other datasets, as a way to evaluate the compatibility between pairs of tasks. Findings. On algorithmic toy tasks, we identify new architectures with dramatic improvements in learning speed, in- and out-of-distribution generalization, and stability across seeds. The new designs prove very task-specific however, and indicate that these tasks require inductive biases very different from those of standard transformers. On code and language modeling datasets, we also find architectures with consistent, yet smaller improvements. These designs transfer much better across datasets and domains (English & computer code). Implications. Our results show that standard transformers are rarely a local optimum in the space of architectures. Simple alternatives can perform much better but sacrifice universality. This suggests that there may be room for improved architectures that better support multiple capabilities simultaneously, such as fluency and robust reasoning.

AI 总结：研究多元时间序列的主动类增量学习，提出TypiCore混合查询策略，结合多种方法评估多种查询策略，揭示基于不确定性和分布感知方法局限性，该策略在多数据集上取得显著改进，减少标注需求。

链接：https://arxiv.org/abs/2607.17632

机构：Budapest University of Technology and Economics（布达佩斯技术与经济大学）； University of Padua（帕多瓦大学）

作者：Gabor Szucs, Samuel Jacsev, Marcell Nemeth, Davide Dalle Pezze, Gian Antonio Susto

英文摘要：Time series data play a pivotal role across numerous domains, including healthcare and manufacturing. In real-world environments, models must cope with distribution shifts over time, a challenge commonly addressed through Continual Learning (CL) techniques. However, existing CL methods face a critical limitation: real-world data streams are rarely fully labeled, making annotation cost a major practical constraint. This paper investigates Active Class-Incremental Learning (ACIL) for multivariate time series, where a model must sequentially learn new classes while selectively querying labels under a fixed annotation budget. We present a systematic evaluation of a wide range of query strategies combined with multiple rehearsal-based approaches, assessing their impact on plasticity, stability, and label efficiency across four benchmark datasets. Our analysis reveals the limitations of uncertainty-based and distribution-aware methods in achieving strong performance under constrained labeling budgets. To address these shortcomings, we propose TypiCore, a novel hybrid query strategy that alternates between typicality-based and diversity-based sample selection across active learning cycles, enabling the construction of memory buffers that are both representative and diverse. Evaluated on the TSCIL benchmark, TypiCore delivers statistically significant improvements over all baselines and matches or surpasses fully supervised continual learning performance on multiple datasets while requiring a fraction of the available labels.

AI 总结：研究特殊活动行人预测问题，采用预训练时间序列基础模型进行零样本概率预测，以SAIL2025事件为案例评估两个模型，为人群管理者明确零样本预测可靠的时机，助力特殊活动人群管理的运营决策。

链接：https://arxiv.org/abs/2607.17758

机构：Delft University of Technology (TU Delft)（代尔夫特理工大学）； German Aerospace Center (DLR)（德国航空航天中心）

作者：Ziteng Li, Yanan Xin, Tina Comes, Serge Hoogendoorn

英文摘要：Managing massive crowds during infrequent special events requires reliable real-time pedestrian-flow forecasting to ensure public safety and operational efficiency. However, supervised forecasting methods face limitations in these contexts due to scarce historical data, heterogeneous data distributions, and short in-event observation windows. To effectively support operational decision-making, forecasts should provide not only accurate point estimates but also informative predictive uncertainty. Probabilistic uncertainty quantification plays a critical role in this aspect, particularly capturing sudden volatility and tail risks. This paper investigates pretrained time series foundation models as a lightweight approach for zero-shot probabilistic forecasting without extensive local retraining. Using decision-oriented metrics tailored to short events, we conduct a comprehensive assessment of two time series foundation models on crowd forecasting, with the SAIL2025 event as a use case. We then distill practical insights for crowd managers, specifying when zero-shot forecasts remain operationally reliable.

AI 总结：研究分布偏移下数据生成问题，提出基于相似度的生成网络SGN，在源数据训练后无需参数更新用于新目标域，通过学习潜在空间和利用目标域小代表性集，实现目标引导数据增强，实验验证其有效性。

链接：https://arxiv.org/abs/2607.18072

机构：National University of Singapore（新加坡国立大学）； Renmin University of China（中国人民大学）； Beijing Institute of Technology（北京理工大学）； Zhejiang University（浙江大学）

作者：Jiaqi Zhu, Xincheng Chen, Yuncheng Wu, Zhaojing Luo, Beng Chin Ooi

英文摘要：Generative models trained on a source domain often produce samples that are poorly aligned with shifted target domains, limiting their effectiveness for target-domain data augmentation. Although target-specific adaptation can reduce this mismatch, it typically requires additional optimization and domain-specific parameters. We propose a Similarity-based Generative Network (SGN), a reusable framework that is trained once on labeled source data and applied to new target domains without parameter updates. SGN learns a latent space structured by label-induced pairwise similarities while preserving reconstructive information through an encoder-decoder architecture. At generation time, a small labeled representative set from the target domain is encoded and combined in the learned latent space, allowing the generated samples to inherit target-specific characteristics while maintaining class consistency. We further analyze the realizability and dimensionality requirements of the proposed similarity structure. Experiments on image and tabular datasets demonstrate the effectiveness of SGN for target-guided data augmentation under source-to-target distribution shifts.

AI 总结：研究分布偏移下类条件覆盖的标签复杂性，指出无标签方法难以兼顾有效性和效率，精确了恢复每类有效性的成本，通过骨架动作识别等案例研究表明相关现象在多模态基准上存在。

链接：https://arxiv.org/abs/2607.18088

机构：University of Washington（华盛顿大学）

作者：Weijia Han, Lisha Qu

英文摘要：Standard evaluation of many recognition systems contains distribution shift by construction, since benchmarks place disjoint conditions in the training and test splits. Under such a shift, split conformal prediction keeps marginal coverage near the nominal level while per-class coverage fails silently: on a real cross-subject skeleton benchmark, marginal coverage stays near ninety percent, the worst action class is covered about seventy percent of the time, and ten of the sixty classes fall below eighty percent coverage. We characterize the cost of restoring per-class validity. First, an impossibility: once the shift acts jointly on the covariates and the labels, the target class-conditional score law is unidentified from source labels and an unlabeled target sample, so no label-free method attains per-class coverage that is at once valid and efficient. Second, we make the cost precise: per-class validity alone needs only a handful of target labels per class, while the label count necessary and sufficient for validity together with per-class efficiency grows as the inverse square of the efficiency tolerance and the logarithm of the number of classes, with matching upper and lower bounds. Third, within the evaluated prediction-powered inference family, even the most favorable use of the classifier's own pseudo-labels on an unbounded unlabeled target pool improves efficiency by at most a small constant factor where coverage collapses. Skeleton action recognition is our real-data case study. A per-class calibration using source labels alone recovers a substantial share of the per-class gap while the shift preserves marginal coverage, and stops helping exactly when marginal coverage itself breaks. Three real shifts of increasing severity trace this boundary, and the same collapse and recovery appears on a natural-image corruption benchmark, beyond any single modality.

AI 总结：研究软标签贝叶斯误差估计器在概率非真实后验时的脆弱性，通过刻画温度缩放对代理的影响，证明相关恒等式并推导封闭形式，在多数据集上验证，给出失真预测，强调代理值与概率产生机制结合才有意义。

链接：https://arxiv.org/abs/2607.18162

机构：Binghamton University (SUNY)（纽约州立大学宾汉姆顿分校）； School of Computing（计算机学院）

作者：Shreyas Pradeepkumar Khandale

英文摘要： The soft-label Bayes-error estimator beta(z) = E[min(z, 1-z)] of Ishida et al. estimates the irreducible error of a binary task directly from probability-valued labels. Recent work by Ushio et al. showed that this estimator is fragile when the probabilities are not the true posterior: even perfectly calibrated soft labels can yield a substantially inaccurate estimate, and they propose isotonic calibration as a consistent remedy. We complement that line of work by characterizing exactly how the most widely used post-hoc calibration map -- temperature scaling -- distorts the proxy. We prove an exact, model-free identity reducing the temperature-scaled proxy to the classifier's margin distribution, from which we obtain (i) strict monotonicity in the temperature and (ii) a continuous bijection from the temperature axis onto the open interval (0, 1/2), so that a fixed classifier -- with fixed decisions and fixed 0-1 error -- can be made to report any proxy value whatsoever. Under a Gaussian model of the logits we further derive a two-parameter closed form for the entire proxy-versus-temperature curve. Across CIFAR-10, Fashion-MNIST, and SVHN (eight binary tasks), the proxy varies by 56x to 980x at constant test error, the closed form reproduces the empirical curve to within 0.018, and the calibration temperature that minimizes the expected calibration error does not coincide with any stable proxy value. Our results give a precise, predictive account of the distortion whose existence motivates calibration-based remedies, and they reinforce the practical recommendation that a proxy value is meaningful only together with the mechanism that produced its probabilities.

AI 总结：研究门级网表中结构可操纵性，定义拓扑驱动分数，用图神经网络制定节点级回归学习该分数，实验评估不同GNN架构效果，组件级和消融分析因素贡献，案例研究揭示注入木马电路结构模式。

链接：https://arxiv.org/abs/2607.16245

作者：Rupesh Raj Karn, Ozgur Sinanoglu

英文摘要：Gate-level netlists exhibit intrinsic structural properties that influence signal propagation independently of functional simulation. We define a topology-driven structural manipulability score that characterizes node-level structural flexibility using path participation, k-core embedding, symmetry, and centrality. Modeling netlists as directed graphs, we formulate node-level regression to learn this topology-derived score using graph neural networks (GNNs). Experiments on ISCAS85 and EPFL benchmarks evaluate how effectively different GNN architectures approximate this metric across held-out circuits, with hierarchical models yielding the most consistent rankings. Component-level and ablation analyses examine the contribution of individual factors. As an illustrative case study, analysis of Trojan-injected circuits using TrustHub templates reveals statistically distinguishable structural patterns, indicating that topology-based scoring provides complementary structural insight.

AI 总结：针对广义学习系统在处理含噪声和多源数据时的不足，提出多视图图嵌入直觉模糊广义学习系统，融合多视图学习、图嵌入和直觉模糊理论，经多方法评估，该系统在多个数据集上表现良好，AUC分数高且性能稳健。

链接：https://arxiv.org/abs/2607.16728

机构：Indian Institute of Technology Ropar（印度理工学院罗帕尔分校）

作者：Yogesh Kumar, Manju, Mudasir Ganaie

英文摘要：The Broad Learning System (BLS) has been widely used for data classification and is based on a layer-by-layer feed-forward structure. However, it gives the same importance to all data points, which reduces its effectiveness on real-world datasets with noise and outliers. In addition, it does not consider the geometric structure of the data and has limitations in handling data from multiple sources. To address these challenges, we propose a Multi-View Graph-Embedded Intuitionistic Fuzzy Broad Learning System (MVGIFBLS) that integrates multi-view learning, graph embedding, and intuitionistic fuzzy theory into the BLS framework. This design enables the model to combine information from multiple sources and learn more discriminative representations. Graph embedding captures the geometric relationships among samples and improves class separation through intrinsic and penalty subspaces based on local Fisher discriminant analysis. Intuitionistic fuzzy theory enhances robustness to noise, while kernel-based neighborhood analysis captures local data structures. We evaluate the proposed framework on several UCI, KEEL, and AwA benchmark datasets using comparative evaluation, Gaussian feature noise analysis, ablation studies, and statistical analysis. The results demonstrate that each component contributes positively to the overall framework and that the proposed MVGIFBLS consistently achieves higher Area Under the Curve (AUC) scores and maintains robust performance under Gaussian feature noise.

AI 总结：本文探讨图神经网络在供应链优化中的潜力，创建大型公共训练数据集，探索适用于节点和网络级预测的GNN架构并分析其与模拟的精度-计算权衡，还概述了基于梯度的拓扑优化等关键研究方向。

链接：https://arxiv.org/abs/2607.16769

机构：School of Mathematics and Computer Science, Indian Institute of Technology Goa（印度理工学院果阿分校数学与计算机科学学院）

作者：Tushar Lone, Neha Karanjkar

英文摘要：Graph Neural Networks (GNNs) have emerged as a powerful, differentiable class of learning models for graph-structured systems. Their ability to generalize across topologies opens the prospect of a surrogate for combined structural and parametric optimization, which classical metamodels cannot offer. Supply chains are a natural target, yet the use of GNN surrogates for supply chain problems is largely unexplored. This paper lays the foundation, presents initial steps, and discusses key research directions. As a foundation, we formulate the problem and create a large public training dataset of programmatically generated supply chain graphs with input parameters and steady-state performance metrics obtained using our SupplyNetPy simulation library. As initial steps, we explore GNN architectures that work well as surrogates for node- and network-level predictions, and analyze their accuracy-compute trade-off against simulation. Most importantly, we outline the exciting directions this opens, namely gradient-based optimization over topology, fast design-space exploration, and sensitivity analysis.

AI 总结：研究通过图同构测试，探讨具有全局注意力的图基础模型对混合整数线性规划的表达能力，证明一类分层图变换器受1-WL测试限制，验证多种图编码器，表明超越其表达能力源于输入编码，提供与编码器无关的诊断方法。

链接：https://arxiv.org/abs/2607.17570

机构：University of Southern California（南加州大学）； USC Information Sciences Institute（南加州大学信息科学研究所）

作者：Md Abrar Jahin, Craig A. Knoblock, Jay Pujara

英文摘要： Graph foundation models (GFMs) with global attention are increasingly used to represent mixed-integer linear programs (MILPs), aiming to capture structure beyond the locality of standard graph neural networks. We study their expressive power through graph isomorphism testing, asking which MILP instances they map to identical representations. We prove that a broad class of hierarchical graph transformers combining global linear attention, edge-weighted cross-attention, and bipartite message passing is bounded by the one-dimensional Weisfeiler-Leman (1-WL) test: under any parameter setting, 1-WL-equivalent MILP graphs receive identical graph embeddings. Our compositional proof shows that each architectural component is a symmetric multiset function and thus preserves 1-WL equivalence. We validate this characterization across ten diverse graph encoders, including Graphormer-, GraphGPS-, Set-Transformer-, and Gasse-style models. Across model capacities, graph scales, and pooling operators, every tested encoder maps 1-WL-equivalent non-isomorphic graph pairs to numerically identical embeddings. Consequently, graph invariants that vary within a 1-WL equivalence class cannot be recovered from these representations. We further show that expressiveness beyond 1-WL arises from input encoding rather than attention: random-walk positional encodings separate the constructed pairs, while additional constructions expose the limits of this remedy. These results characterize the expressive power of global-attention GFMs and provide an encoder-agnostic diagnostic for detecting 1-WL-induced representation equivalence.

AI 总结：研究等变图神经网络中传统张量积卷积实现的问题，提出通过重新关联操作优化的流形公式，在Sobek中实现后经评估，相比传统方法有显著速度提升、内存减少，能处理更大工作负载。

链接：https://arxiv.org/abs/2607.18074

机构：Cortex Discovery（皮质发现公司）

作者：Vladimir Chorošajev, Cédric Bény

英文摘要：Equivariant graph neural networks repeatedly apply edge-conditioned tensor-product convolutions over graph edges. Conventional implementations materialize edge-specific weights, messages, and adjoints, causing tensor-product workspace and memory traffic to grow rapidly with graph size and operator width. This limits feasible workloads and can prevent larger problems from fully utilizing the GPU. We show that these edge-sized intermediates are artifacts of the execution schedule, not requirements of the equivariant operator. By reassociating radial projection, spherical-harmonic coupling, and graph aggregation, edge-local products can be consumed directly into bounded receiver-side state. The resulting streaming formulation preserves fully connected multiplicity mixing and extends through forward, backward, and double backward. We implement this formulation in Sobek, a generated-CUDA backend, and evaluate it across edge-scaling regimes and varied feature structures. Across two operator families and all three differentiation orders, Sobek is faster in all 75 capacity-matched comparisons, with speedups ranging from $1.2\times$ to $49.7\times$, and reduces peak allocated memory by up to 99\%. It also executes workloads up to two orders of magnitude beyond OpenEquivariance's capacity while retaining near-peak throughput. These results show that edge-scaled tensor-product workspace is a property of the conventional schedule, not of equivariant convolution itself.

AI 总结：研究针对大语言模型中LoRA方法的不足，提出SOS-LoRA，将秩更新重参数化为多个静态低秩专家之和，通过分解总秩、固定多尺度缩放及正交初始化等，实现性能提升且可完全合并，实验证明其优于基线和变体。

链接：https://arxiv.org/abs/2607.16252

机构：School of Artificial Intelligence, Jilin University（吉林大学人工智能学院）； Engineering Research Center of Knowledge-Driven Human-Machine Intelligence, MOE, China（教育部知识驱动的人机智能工程研究中心）； International Center of Future Science, Jilin University（吉林大学未来科学国际合作中心）

作者：Yupeng Chang, Yuan Wu, Yi Chang

英文摘要：Low-Rank Adaptation (LoRA) is a widely used parameter-efficient fine-tuning (PEFT) method for large language models. Under a fixed rank budget, LoRA parameterizes each adapted weight through a single low-dimensional input-side pathway, which may couple heterogeneous behaviors through shared input directions and induce interference during optimization. We propose Static Orthogonal Subspace LoRA (SOS-LoRA), a drop-in extension that reparameterizes a rank-rtot update as a sum of K static (always-on, non-routed) low-rank experts. SOS-LoRA (i) decomposes the total rank across experts, (ii) applies a fixed multi-scale scaling scheme to encourage scale-separated optimization dynamics, and (iii) promotes diverse input-side directions via cross-expert orthogonal initialization and a lightweight regularizer. SOS-LoRA remains fully mergeable, adding no inference-time parameters or latency after merging. Experiments on reasoning and knowledge-intensive benchmarks (Llama 2/3), encoder-based NLU (GLUE), and math reasoning (GSM8K/MATH) show consistent gains over matched-budget LoRA baselines and recent variants. Code is available at this https URL.

AI 总结：研究探讨人工记忆中跨域重组，通过在两个系统实现重组重播机制，发现跨域巩固创造价值，域内排练不然，符号系统跨域连接提升显著，神经系统在特定子任务有提高，验证了相关预测，表明巩固是为发现而非记忆。

链接：https://arxiv.org/abs/2607.16256

机构：University of Chicago（芝加哥大学）； Santa Fe Institute（圣塔菲研究所）； Stanford University（斯坦福大学）

作者：Oliver Zahn, James Evans, David Eagleman

英文摘要： Dreams splice together people, places, and times that never met. Neuroscience suggests this recombination is not noise, but a function driving insight and creative discovery. This reframes memory consolidation: rather than merely defending against forgetting, its measurable value lies in recombining knowledge across experiences that have not yet co-occurred. We test this directly by isolating the recombinatory-replay mechanism and implementing it in two architecturally unrelated systems: a LoRA fine-tuning pipeline (DREAMS) and a symbolic engine replaying structured knowledge objects (SAPIENCE). Both systems converge on the same finding: cross-domain consolidation creates value, while within-domain rehearsal does not. The symbolic arm surfaces novel cross-domain connections at 85.7%, a +21 percentage point (pp) gain over baseline. The neural arm improves overall by +5.64 pp, but on subtasks explicitly requiring cross-domain transfer (like unseen math reasoning on GSM8K), gains reach +14.5 pp. This effect is a genuine property of the weights--not a prompt artifact--as prepending the same material in-context to a 671B-parameter model actually reverses the gain. We validate this prediction against documented discoveries across 50,000 real papers and state a falsifiable hippocampal-recording prediction to distinguish recombination from rehearsal. Ultimately, this principle is substrate-general, tracking real discovery at scale. Reading the literature teaches a model to recall what it has seen, but producing discovery requires a separate offline phase that recombines knowledge across domains--the computational analog of dreaming. Consolidation is not for remembering, but for discovering.

AI 总结：研究分层域泛化问题，从有限观察区域外推到整个实例空间，以任意域层次结构取代独立同分布采样，揭示核心障碍在于训练/测试域划分，表明现代泛化理论应将域结构视为关键因素。

链接：https://arxiv.org/abs/2607.16528

机构：Toyota Technological Institute at Chicago（芝加哥丰田技术学院）； University of Waterloo（滑铁卢大学）

作者：Chenxiao Yang, Zhiyuan Li, Shai Ben-David, Nathan Srebro

英文摘要：We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain hierarchies. We show that the central obstruction is not only the complexity of the hypothesis class, but the train/test domain partition through which evidence is revealed. In particular, no matter how small the class or how large the training size, some partition makes generalization fail for some target. These results suggest that modern generalization theory must treat domain structure as a first-class object.

AI 总结：研究多任务学习中负迁移与容量、冗余的关系，通过容量 - 冗余恒等式及相关结果，如聚类差距分解和梯度 - TC 桥梁，证明聚类 LoRA 可降低残余耦合，优于随机划分，有显著收益。

链接：https://arxiv.org/abs/2607.16554

机构：Harvard Medical School（哈佛医学院）

作者：Asif Khan

英文摘要：In multi-task learning (MTL) negative transfer is often considered as an optimization artifact, but it can also be viewed as a consequence of limited shared capacity and weak task redundancy. We investigate this effect through a Capacity--Redundancy (CR) identity that decomposes the sum of per-task predictive informations into joint predictive information that includes label redundancy defined via total correlation (TC), and a residual coupling term that quantifies interference left unresolved by the shared representation. Additionally, we show two key results: (i) a clustering-gap decomposition that gives a necessary and sufficient condition for clustered sharing to outperform global sharing, and (ii) a gradient--TC bridge in a Gaussian multi-task model that formally justifies gradient cosine similarity as a proxy for redundancy ordering. Empirically, we estimate the residual coupling $\Delta$ from validation residual correlations, showing that clustered LoRA substantially reduces $\widehat{\Delta}$, outperforms size-matched random partitions, and results in statistically significant gains with multi-seed confidence intervals.

AI 总结：研究针对多模态模型训练障碍，提出MultiLoReFT框架，通过低秩表示微调，将低秩适应扩展到多模态，学习可解释投影子空间解耦信息，能在多模态预测时揭示信息分布。

链接：https://arxiv.org/abs/2607.16789

作者：Sana Tonekaboni, Viktoria Schuster, Caroline Uhler

英文摘要：Real-world perception and decision making are inherently multimodal, integrating complementary signals across modalities. However, training multimodal models faces two main obstacles. First, collecting large-scale, well-aligned paired multimodal datasets is often impractical, making end-to-end multimodal training difficult. Second, existing multimodal representations frequently entangle information shared across modalities with modality-specific information, hindering interpretability and control. We introduce MultiLoReFT, an efficient and scalable low-rank representation fine-tuning framework for multimodal learning with pretrained unimodal models. MultiLoReFT extends low-rank adaptation to the multimodal setting and learns interpretable projection subspaces that decouple shared and modality-specific information. Across simulated and real-world benchmarks, it produces representations that support multimodal prediction while explicitly revealing how shared and modality-specific information is distributed across modalities.

AI 总结：研究训练后变压器中局部任务适应，通过多任务LoRA操作点测量相关属性，发现一阶变化可预测，成对结构脆弱，如两更新顺序敏感、任务梯度子空间旋转等，还给出连续任务梯度步骤相关项及预测与起始尺度情况。

链接：https://arxiv.org/abs/2607.16821

机构：DAIMLD（数据挖掘与机器学习系）； St. Petersburg Department of the Steklov Institute of Mathematics（斯捷克洛夫数学研究所圣彼得堡分部）； St. Petersburg State University（圣彼得堡国立大学）

作者：Irina Piontkovskaia, Sergey Nikolenko

英文摘要： Task arithmetic, sequential fine-tuning, activation steering, and first-order random search all operate through relatively small perturbations around an already trained checkpoint, and they rely on different local approximations: individual perturbations should be first-order predictable, task updates should compose with controlled interference, useful tangent structure should be stable and possible to estimate, and weight edits should have counterparts in representation space. We measure 8 such properties with the same harness around a multitask LoRA operating point, on 9 transformers (82M-7B), with a prospectively registered property list, thresholds, and test split. We find a shared one-direction validity window up to the tested scale $10^{-2}$, but no universal radius for pairwise composition or update ordering. Along individual directions, changes of the probe loss remain first-order predictable throughout the grid: a perturbation's effect on the loss is essentially its projection onto the gradient, which is also what makes local random search work. Pairwise structure, however, proves to be far more fragile: on over a third of the measured (model, task pair) combinations, two-update order sensitivity sets in strictly inside that window; task-gradient subspaces rotate within tens of steps; additivity under our fixed activation probe fails at full task-vector scale on several models, including both held-out 7B models; and no model median passes the registered global mean-vector weight-to-steering correspondence bar. For two sequential task-gradient steps, the leading order-dependent term is the Lie bracket $H_B\textbf{g}_A-H_A\textbf{g}_B$; its normalized prediction $c(\eta)=\eta\kappa+O(\eta^2)$ tracks the measured defect at median ratio 1.002, while the onset scale $\eta^\dagger\approx0.10/\kappa$ spans three orders of magnitude across models and task pairs.

AI 总结：研究提出ChebyMA参数高效自适应方法，通过切比雪夫多项式基多表面叠加采用权重矩阵。理论上证明其收敛性与表达优势，实验表明在CIFAR和自然语言处理数据集上，该方法比其他方法参数-精度表现更优，验证了理论基础。

链接：https://arxiv.org/abs/2607.17377

作者：Jiawen Li

英文摘要：The paper presents a new parameter-efficient adaptation method called ChebyMA (Chebyshev Manifold Adaptation). ChebyMA adopts weight matrices through a multi-surface superposition of Chebyshev polynomial bases evaluated on learnable coordinates and combined via trainable coefficient matrices, replacing standard linear projections with highly expressive continuous function approximation. Theoretically, we establish an Approximation Expressivity Theorem, proving from the perspective of function approximation theory that single-manifold ChebyMA guarantees convergence in Frobenius norm error of reconstruction. Besides, drawing on Kolmogorov $n$-width intuition, we demonstrate the expressive advantages of multi-manifold superposition ($S > 1$) in decoupling high-dimensional complex features. Experimental results on Computer Vision CIFAR datasets(CIFAR-10, CIFAR-100)\cite{CIFAR} and Natural Language Processing (AG News, SST-2) datasets demonstrate that ChebyMA consistently achieves a superior parameter-accuracy Pareto front compared to standard full-parameter fine-tuning, LoRA\cite{LoRA}, TLoRA\cite{TLoRA}, and StelLA\cite{StelLA}. ChebyMA significantly outperforms other tested methods in tested datasets, validating its solid theoretical foundation for generality with purely vectorized computations.

AI 总结：研究针对医疗保健领域特定任务零样本模型问题推出检索增强可解释学习（RAIL）框架，通过检索相关任务并传递结构生成新预测器，概率公式提供不确定性支持可靠性感知部署，在临床程序预测任务中性能可靠，还提升模型透明度。

链接：https://arxiv.org/abs/2607.17508

机构：Carnegie Mellon University（卡内基梅隆大学）； University of Wisconsin–Madison（威斯康星大学麦迪逊分校）； Mohamed bin Zayed University of AI（穆罕默德·本·扎耶德人工智能大学）； GenBio AI（基因生物人工智能公司）； Intel（英特尔公司）

作者：Sazan Mahbub, Caleb Ellington, Zhiyuan Li, Yixin Yang, Souvik Kundu, Ben Lengerich, Eric P. Xing

英文摘要：We introduce Retrieval-Augmented Interpretable Learning (RAIL), a probabilistic meta-learning framework for zero-shot generation of task-specific interpretable models that synthesizes coefficient-space structure from natural-language task descriptions and a memory of previously learned task-specific predictors. RAIL retrieves related source tasks, transfers structure through coefficient space, and generates a new predictor in the original diagnostic-feature space, enabling zero-shot and few-shot clinical procedure prediction with feature-level explanations. Its probabilistic formulation provides uncertainty over retrieval, model coefficients, and predictions, supporting reliability-aware deployment: uncertain predictions or unstable explanations can be flagged for additional clinical review rather than treated as automatic decisions. This makes RAIL particularly suited for healthcare settings, where prediction tasks are highly long-tailed, new clinical targets arise frequently, and models must remain inspectable, uncertainty-aware, and compatible with human oversight. Across long-tailed clinical procedure prediction tasks, RAIL maintains reliable performance across data-availability regimes: it achieves 73.4% accuracy in the held-out zero-shot settings, where no supervised task-specific model can be trained, and remains near 73.2% accuracy in the extreme few-shot regime with only 2-4 examples, where supervised task-specific models perform close to chance. RAIL further benefits from clinically informed task representations and yields retrieval, uncertainty, and coefficient-level diagnostics that make model behavior more transparent. These results suggest a path toward scalable clinical prediction systems that can adapt to new tasks while preserving interpretability and reliability.

AI 总结：研究针对LoRA优化问题，提出PoLoRA优化器，由乘积感知谱更新方向、曲率预处理及幅度规则构成。在多模型指令调优数据集上评估，结果显示它步数减少，开销增加少，对学习率不敏感，最优学习率稳定。

链接：https://arxiv.org/abs/2607.17620

机构：Center for Computational Mathematics, Flatiron Institute（计算数学中心，熨斗研究院）

作者：Nikhil Ghosh, Tetiana Parshakova, Robert M. Gower

英文摘要：Low-rank adaptation (LoRA) makes finetuning large language models cheaper by adding to each weight matrix a trainable low-rank update parameterized as the product of two matrices. These matrices are usually trained with Adam, which treats them as a single flat vector of parameters and ignores both the matrix and product structure of LoRA. Applying a matrix-aware optimizer such as Muon to each factor does not consistently improve over Adam, and neither do the product-aware Muon variants proposed in concurrent works. To realize consistent gains, we introduce PoLoRA, a Preconditioned Orthogonalized LoRA optimizer built from three ingredients: a product-aware spectral update direction, curvature preconditioning derived from controlling the per-sample loss change, and a magnitude rule that controls the sizes of both the factor and merged updates. We evaluate PoLoRA on instruction-tuning datasets for code and math across models from 1B to 8B parameters, and find that it reaches the final held-out loss achieved by tuned Adam in 1.2-1.7 times fewer steps, while adding at most 3% per-step overhead. Compared to Adam, PoLoRA is also less sensitive to the learning rate, and its optimal learning rate is stable across ranks.

AI 总结：研究无监督图域适应问题，提出源节点影响剪枝（SNIP）方法，通过量化结构差异为源节点打分并过滤低影响分数节点，构建精炼“子源”图，实验证明该方法优于基线，验证了选择性节点利用的优越性。

链接：https://arxiv.org/abs/2607.17668

机构：School of Computer Science and Technology / School of Artificial Intelligence, China University of Mining and Technology（中国矿业大学计算机科学与技术学院/人工智能学院）； Mine Digitization Engineering Research Center of the Ministry of Education（教育部矿山数字化工程研究中心）； Jiangsu Provincial Industrial Technology Engineering Center for Intelligent Sensing and Emergency IoT in Underground Space（江苏省地下空间智能传感与应急物联网产业技术工程中心）； School of Electrical Engineering and Computer Science, University of Ottawa（渥太华大学电气工程与计算机科学学院）

作者：Ridong Han, Yawen Shen, Zhongnian Li, Tongfeng Sun, Xinzheng Xu, Abdulmotaleb El Saddik

英文摘要：Unsupervised Graph Domain Adaptation (UGDA) aims to facilitate knowledge transfer from a labeled source graph to an unlabeled target graph by mitigating cross-domain distribution shifts. Existing methods primarily focus on node-level feature alignment in latent spaces, relying on the implicit assumption that all source nodes contribute positively to the alignment. However, this assumption often fails because a node's semantic information is intrinsically coupled with its topological graph structure. Due to structural shifts, source nodes with severe structural deviations (e.g., structural outliers) lack semantic counterparts in the target graph, and forcing alignment on them introduces severe noise and causes negative transfer. To bridge this gap, we argue that selective source node utilization is superior to full-graph training, thereby shifting the research paradigm from feature-level alignment to data-level refinement. To this end, we propose Source Node Influence Pruning (SNIP), a novel model-agnostic, data-centric refinement framework. Specifically, SNIP quantifies the structural discrepancy between individual source nodes and the target domain by integrating multiple centrality measures, assigning each source node an influence score. A rank-based normalization mechanism is further employed to eliminate scale variations across different measures, allowing SNIP to effectively identify and filter out structurally incompatible nodes with low influence scores. As a plug-and-play method, SNIP constructs a refined "sub-source" graph that is inherently more beneficial for subsequent alignment. Comprehensive experiments across eight transfer scenarios on five real-world datasets demonstrate that SNIP consistently outperforms competitive baselines and significantly enhances adaptation performance, validating the superiority of selective node utilization over full-graph training.

AI 总结：研究将流形约束超连接作为参数高效微调新方法，包裹冻结主干。发现其与预训练设置作用不同，固定残差混合矩阵有益。独立时不比LoRA优，但mHC+LoRA组合在特定规模下有优势，确定残差路由是有前景的新型PEFT轴。

链接：https://arxiv.org/abs/2607.18130

机构：University of Amsterdam（阿姆斯特丹大学）

作者：Valentijn Oldenburg, Floris de Kam, Bente Zuijdam, Lieve Eberson, Nicky van Zutphen, Stef de Wildt, Ivo Verhoeven

英文摘要：Most parameter-efficient finetuning (PEFT) methods adapt weights or activations, thus leaving one of the key Transformer components unchanged: residual connections. This paper investigates Manifold-Constrained Hyper-Connections (mHC), a generalisation of residual connections, as a novel PEFT approach, wrapping frozen OLMo-2 backbones with learned residual routing modules. We find that mHC can finetune frozen Transformers, but that its role differs fundamentally from the original pre-training setting: in finetuning, fixing the residual mixing matrix to identity often improves performance. As a standalone PEFT method, mHC does not consistently outperform LoRA. However, at matched trainable parameter budgets, mHC+LoRA combinations improve language-modelling loss and show task-dependent benchmark gains at both 1B and 7B scale. Overall, our results identify residual routing as a distinct and promising novel PEFT axis.

AI 总结：研究针对数字孪生中概念漂移致模型退化问题，提出集成Fisher分数漂移检测器、LoRA及Mann-Whitney U检验的自适应框架，能监测模型置信度、触发微调并统计验证性能提升，经案例研究成功检测变化并恢复相关性能，维持数字孪生可信度。

链接：https://arxiv.org/abs/2607.18164

机构：Northwestern University（西北大学）

作者：Yi-Ping Chen, Ying-Kuan Tsai, Vispi Karkaria, Seul Lee, Daniel Apley, Wei Chen

英文摘要：Digital Twins rely on surrogate models to mirror physical systems in real time, yet these models can degrade as operating conditions evolve, a phenomenon known as concept drift. Maintaining surrogate fidelity under drift, particularly when models must also capture aleatoric uncertainty, remains an open challenge. Existing adaptive frameworks lack principled mechanisms for detecting when updates are needed, for efficiently adapting models from limited streaming data, and for certifying that updates genuinely improve predictive performance. Here we present an adaptive Digital Twin framework that integrates a Fisher score--based multivariate drift detector, Low-Rank Adaptation (LoRA) for parameter-efficient continual learning, and a Mann--Whitney $U$ test for online statistical validation. The framework monitors surrogate-model confidence via Fisher score vectors, triggers targeted fine-tuning of fewer than 1% of model parameters upon drift detection, and statistically certifies predictive improvement before deploying the updated surrogate. Applied to a stochastic linear system and a directed energy deposition additive manufacturing process as case studies, the framework successfully detects distributional shifts with short delays and restores both predictive accuracy and uncertainty quantification under abrupt and incremental drift. These results establish a statistically rigorous and computationally tractable pathway for sustaining the trustworthiness of neural-network--based Digital Twins throughout their operational life cycle.

AI 总结：研究针对基于多组学数据预测乳腺癌雌激素受体（ER）状态，对随机森林等经典机器学习模型进行基准测试，采用严格实验框架，结果显示RNA表达预测信号最强，多组学整合有改进，随机森林性能最佳，还选出相关基因支持模型有效性。

链接：https://arxiv.org/abs/2607.16250

作者：Priyanka Paudel, Madan Baduwal

AI 总结：研究旨在学习时空基础模型，提出NeoST，通过在程序生成的合成系统上预训练，引入可扩展语料库、潜在空间推理架构和目标，实验证明其在多样真实世界时空系统中性能优越，有长期稳定性和推理效率。

链接：https://arxiv.org/abs/2607.16251

机构：Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））； National University of Singapore（新加坡国立大学）； Hong Kong University of Science and Technology（香港科技大学）； Hong Kong Polytechnic University（香港理工大学）

作者：Yutong Feng, Shiyuan Piao, Yutong Xia, Xu Liu, Wenqi Fan, Fugee Tsung, See-Kiong Ng, Yuxuan Liang

AI 总结：研究量化语言模型基准测试中排名不确定性问题，通过汇总成对假设检验来实现，分析了知识评估基准MMLU的不确定性来源并展示如何修改假设检验，指出MMLU各主题排名变异性大，比较模型时应考虑。

链接：https://arxiv.org/abs/2607.16259

作者：Bitya Neuhof, Yuval Benjamini

AI 总结：研究针对自适应优化器在不同训练阶段更新规则相同的问题，提出PsiLogic优化器，通过双指数移动平均控制主动抵消项，经FairBench基准测试，在多个领域表现优异，还发布开源实现等支持验证。

链接：https://arxiv.org/abs/2607.16268

作者：Ali Sultonov

AI 总结：针对噪声标签学习问题，提出从单变量基函数构建鲁棒多类损失的框架，开发两种方案并分析其性质，推导理论标准，所提损失在多噪声设置下经实验验证有竞争力或更优。

链接：https://arxiv.org/abs/2607.16768

机构：School of Mathematics, Harbin Institute of Technology（哈尔滨工业大学数学学院）； School of Earth and Space Sciences, Peking University（北京大学地球与空间科学学院）

作者：Peng Hu, Jianwei Ma

AI 总结：研究提出两个全文科学记忆基准测试PAIM和PTr，评估八个内存/检索系统，发现内存排行榜受多种因素影响，如摄取粒度等。结果显示不同系统表现因条件而异，还表明大语言模型评判排名与人工一致，应将科学记忆评估为预算上下文恢复并开源相关资源。

链接：https://arxiv.org/abs/2607.16848

机构：Quantellence Research（Quantellence研究公司）； St. Petersburg Department of the Steklov Institute of Mathematics（斯捷克洛夫数学研究所圣彼得堡分部）； St. Petersburg State University（圣彼得堡国立大学）

作者：Maksim Sheverev, David Finkelstein, Sergey Nikolenko

AI 总结：研究RAG系统向量检索层挑战，介绍基于TurboQuant的TurboVec开源向量索引。通过实验对比，表明TurboQuant在内存和召回率上有优势，TurboVec部署效果好，码本无关设计提升隐私性，不过存在单一数据集评估等局限。

链接：https://arxiv.org/abs/2607.16973

作者：Navnit Shukla, Kamal Pandey, Omsankar Tiwari

AI 总结：研究在误导性证据环境下深度研究代理的表现，引入DRNOISE基准，每个任务含正确答案及冲突文档，涵盖多类证据操作。测试发现代理存在验证惰性，通用提示可缩小差距，强调可靠深度研究需积极协调主张与证据。

链接：https://arxiv.org/abs/2607.17291

机构：Hong Kong Baptist University（香港浸会大学）； University of Science and Technology of China（中国科学技术大学）； The Hong Kong University of Science and Technology（香港科技大学）

作者：Jun Nie, Zhiqin Yang, Zhenheng Tang, Yonggang Zhang, Xiaowen Chu, Xinmei Tian, Bo Han

AI 总结：研究将预训练语言模型响应分布分解为策略条件表示的问题，提出潜在变量分解方法，引入新变分目标，通过多策略算法任务基准验证了该方法能恢复潜在代码并保留基础模型响应分布。

链接：https://arxiv.org/abs/2607.17674

作者：Awni Altabaa, John Lafferty

AI 总结：介绍WC2026-Agents基准和数据集，用于评估大语言模型作为世界杯预测代理。四个前沿模型对104场比赛运行相同循环，与博彩市场数据配对。揭示模型预测虽有相同首选，但在决策质量等方面差异大，可衡量校准等方面。

链接：https://arxiv.org/abs/2607.17765

机构：University of Memphis（孟菲斯大学）； QuantaInsight（量子洞察）

作者：Jiacheng Ding, Cong Guo, Jason Xu

AI 总结：研究TabPFN在复杂表格几何上的内部行为，用曲折持久同调将其层表示视为点云，构建合成表格任务基准。发现其内部表示几何拓扑与数据集级可靠性相关，复杂几何有双重拓扑特征，相关描述符与多种指标有关，能诊断可靠性及运行状态。

链接：https://arxiv.org/abs/2607.17962

机构：TD Bank（道明银行）

作者：James Hu, Mahdi Ghelichi

AI 总结：针对GUI代理轨迹合成中缺乏高覆盖率长视野轨迹的问题，提出SEE两阶段数据合成框架，通过高效探索和基于图的合成，产生可重复可解释数据，避免虚假循环，提升代理任务成功率和泛化能力，还将发布代码和数据集。

链接：https://arxiv.org/abs/2607.18046

机构：Harbin Institute of Technology, Weihai（哈尔滨工业大学（威海））； Harbin Institute of Technology (Weihai) Qingdao Research Institute（哈尔滨工业大学（威海）青岛研究院）

作者：Zhuohang Fan, Beichen Zhang, Yuanfa Li, Changqiao Wu, Wei Liu, Jian Luan, Weigang Zhang

AI 总结：研究全正矩阵能否通过特征多项式最高阶系数区分，采用神经网络分类器等方法，利用多族矩阵构建数据集，发现特定系数有判别力，分离呈非线性，不同族有不同椭球特征并据此提出猜想。

链接：https://arxiv.org/abs/2607.18148

机构：Institute of Informatics, Federal University of Rio Grande do Sul（南里奥格兰德联邦大学信息学院）； Institute of Mathematics and Statistics, Federal University of Rio Grande do Sul（南里奥格兰德联邦大学数学与统计学院）

作者：Tiago Closs, Leandro Farina

AI 总结：为加速可穿戴健康领域开放科学，发布OpenMHC这一最大最全的可穿戴健康数据集及模型开源实现，引入统一开放基准，对多种模型进行测试，通过大规模开源数据、代码和模型权重，推动可穿戴健康AI研究发展。

链接：https://arxiv.org/abs/2607.16235

作者：Narayan Schuetz, Yuze Bai, Lianggang Pan, Edgar Eggert, Favour Nerrise, Juan Delgado-SanMartin, Max Rosenblattl, Milana Gurbanova, Mohammad Asadi, Anders Johnson, Paul Schmiedmayer, Dennis Wang, Allan Lawrie, Daniel Seung Kim, Xin Liu, Akshay Paruchuri, Ehsan Adeli, Euan Ashley, Kelly W. Zhang

AI 总结：研究材料喷射中液滴演化预测难题，提出DiffARFNO两阶段框架，结合自回归傅里叶 - MIONet与DDIM校正器，经实验验证该方法显著优于现有模型，能为长期预测提供高保真结果。

链接：https://arxiv.org/abs/2607.16238

机构：New Jersey Institute of Technology（新泽西理工学院）； University of Georgia（佐治亚大学）； University at Buffalo（纽约州立大学布法罗分校）； Hanyang University（汉阳大学）； Northeastern University（东北大学）

作者：Jinghao Cao, Minsung Kang, Hongyue Sun, Chi Zhou, Jihoon Chung, Xubo Yue, Sanchoy Das, Bo Shen

AI 总结：研究针对多模态小语言模型现场部署的鲁棒性评估问题，开发RobustMAD基准测试，通过多种开放式查询评估模型，发现最佳模型虽有潜力但仍存差距及三种失败模式，为下一代多模态工业检测助手设计提供指导。

链接：https://arxiv.org/abs/2607.16243

机构：Singapore University of Technology and Design（新加坡科技设计大学）； Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局）； Nanyang Technological University（南洋理工大学）； Chongqing University（重庆大学）

作者：Anushiya Arunan, Xin Li, Yan Qin, U-Xuan Tan, Nhu Khue Vuong, Xiaoli Li, Chau Yuen

AI 总结：研究用高斯混合模型检测异常与分布漂移，明确三个实际选择并评估。通过自动选择混合成分数量、负对数似然评分及扩展模型检测漂移，在多基准测试中该模型有竞争力且可解释，还对比了其他方法。

链接：https://arxiv.org/abs/2607.16811

作者：Behnam Asadi

AI 总结：针对多元时间序列预测中捕捉全局与局部依赖关系的难题，提出HyBDM多尺度混合模型，通过两个专家分别建模全局模式与局部变化，结合多尺度补丁器和路由器，实验证明其在预测准确性和计算效率上优于现有方法。

链接：https://arxiv.org/abs/2607.16882

作者：Wenqiang Ma, Chen Cheng, Xue Cheng, Jiarui Ye

AI 总结：研究针对区域气候预测中时间序列基础模型的挑战，提出残差引导多分辨率细化框架RGMR，无需更新主干参数，应用于干旱预测时能降低测试集MSE，为部署冻结的TSFM提供实用途径。

链接：https://arxiv.org/abs/2607.17507

作者：Wentao Gao, Jiuyong Li, Lin Liu, Thuc Duy Le, Jixue Liu, Yanchang Zhao, Yun Chen

AI 总结：研究针对时间序列基础模型应用于区域干旱预测的挑战，提出轻量级黑盒适应框架，通过SMR²和MBB两个包装器在推理时增强模型，在南澳大利亚站点预测中提升性能，降低MSE，实现资源受限区域的实际部署。

链接：https://arxiv.org/abs/2607.17511

机构：Adelaide University（阿德莱德大学）； CSIRO（联邦科学与工业研究组织）

作者：Wentao Gao, Jiuyong Li, Lin Liu, Thuc Duy Le, Jixue Liu, Yanchang Zhao, Yun Chen

AI 总结：研究针对蛋白质-配体结合亲和力预测中引擎结果不一致问题，提出RELIABLE-BA框架，通过三步实现多引擎预测，经实验验证其能提升预测竞争力和不确定性校准，可筛选高置信对减少误差，是首个结合证据融合与上下文可靠性的预测框架。

链接：https://arxiv.org/abs/2607.17601

机构：University of Southern California（南加州大学）； University of California, Berkeley（加州大学伯克利分校）

作者：Yongchan Hong, Defu Cao, Wenjin Liu, Thomas Ku, Jordy Homing Lam, Emily Nguyen, Willie Neiswanger, Vsevolod Katritch, Yan Liu

AI 总结：研究针对物理信息神经网络求解偏微分方程时的局限，提出FlashPDE库，用可微Triton内核取代PyTorch有限差分执行，集成多种功能。实验表明其能大幅减少内存使用和内核启动次数，实现加速，为PDE求解与GPU计算搭建高效执行层。

链接：https://arxiv.org/abs/2607.18020

机构：School of Mathematical Sciences, Beijing Normal University（北京师范大学数学科学学院）； Beijing University of Posts and Telecommunications（北京邮电大学）

作者：Peiyu Zang, Bosen Xie, Ruoxiang Xu, Yongqiang Cai

AI 总结：研究如何准确求解任意几何形状和各种网格上的偏微分方程，提出自适应曼巴神经算子，通过构建竹内 - 马尔姆奎斯特系统集成再生核，在多个领域的基准问题上，其相对\(L^2\)误差优于现有求解器，为神经算子框架设计提供新范例。

链接：https://arxiv.org/abs/2607.18043

作者：Zeyuan Song, Zheyu Jiang

AI 总结：针对文档解析中OCR工具选择难题，尤其是标签稀缺情况，提出无标注评估框架DocOCR-Eval。它采用三阶段校正和排序策略，通过跨多个MLLM聚合改善与基于标注排名的对齐，能在现实有限标签设置中实现可靠OCR工具选择并提供实用指导。

链接：https://arxiv.org/abs/2607.16203

作者：Zihan Xu, Puzhen Wu, Lawrence Chun Man Lau, Wei Liu, Sirui Li, Yifan Peng, Yihao Ding

AI 总结：介绍ARGO智能眼镜平台，通过硬件、固件和人工智能整体协同设计，利用STM32N6微控制器及集成神经处理单元实现设备端机器学习，以优化的YOLOv11模型识别障碍物，展示了高性能、隐私保护辅助设备可行性。

链接：https://arxiv.org/abs/2607.16222

机构：EssilorLuxottica（依视路陆逊梯卡集团）； Politecnico di Milano（米兰理工大学）

作者：Andrea Giudici, Christian Veronesi, Pietro Bartoli, Mario Caliò, Aurelio Teliti, Giacomo Gervasoni, Diana Trojaniello, Franco Zappa

AI 总结：综述聚焦于网络防御中LLM遗忘问题，探讨其面临风险，核心问题是现有方法能否真的去除知识。主要关注基于梯度的方法，因其与现有训练管道兼容且可扩展，从多方面审视LLM遗忘，为解决相关问题提供参考。

链接：https://arxiv.org/abs/2607.16227

机构：Manipal Institute of Technology, Manipal Academy of Higher Education（马尼帕尔理工学院，马尼帕尔高等教育学院）； San José State University（圣何塞州立大学）

作者：Ruppikha Sree Shankar, Abhishek Bhardwaj, Arnav Doshi, Anusri Nagarajan, Troy Paulus Asia, Saptarshi Sengupta

AI 总结：研究张量内核正确性测试的容差校准问题，通过挖掘测试用例误差分布提出新容差，经实验，对特定错误变体校准容差提高了错误检测召回率，虽有少量误报增加，但整体提升了检测效果。

链接：https://arxiv.org/abs/2607.16228

机构：Arizona State University（亚利桑那州立大学）

作者：Dipankar Sarkar

AI 总结：研究电子商务中预订单运输成本估计问题，提出受生产启发的多阶段框架RouteCost，将其分解为多步骤，通过路线加权期望公式汇总成本估计，在大量订单等数据上提高了预测质量和校准，保留了路线级可解释性。

链接：https://arxiv.org/abs/2607.16230

机构：Northeastern University（东北大学）

作者：Xianling Zeng, Zihan Yu, Sichen Zhao, Yalun Qi, Zhiming Xue

AI 总结：研究噪声标签学习中记忆驱动过拟合问题，核心方法是用OrthoGrad去除权重梯度中与当前权重向量平行的分量，在MNIST小数据及CIFAR-10 ResNet-18实验中验证其能改变记忆轨迹，但依赖数据情况，可作学习动态诊断。

链接：https://arxiv.org/abs/2607.16231

作者：Richard Mai

AI 总结：该研究针对统计学习算法推断人类偏好的挑战，提出‘从权重到文字’方法，输入选择问题数据集，自动发现偏好维度，解决欠定和不透明问题，经多领域展示及人体实验验证，能提高偏好模型预测准确性，获参与者认可。

链接：https://arxiv.org/abs/2607.16232

机构：MIT（麻省理工学院）

作者：Zachary Wojtowicz, Ayush Nayak, Jacob Andreas

AI 总结：针对整合异质模态进行癌症亚型分类和生存预测的挑战，提出令牌级跨模态变换器及对比多任务学习方法，克服现有方法在模态交互、融合方式及目标优化上的局限。

链接： https://arxiv.org/abs/2607.16233

机构：Jiangxi Arts & Ceramics Technology Institute（江西艺术陶瓷科技职业学院）； Mokwon University（木浦大学）

作者：Suxing Liu Byungwon Min

AI 总结：研究时间序列模型可解释性方法中基于边际影响的归因方法失效问题，指出主要因计算不匹配，定义DAG忠实性，发现标准归因方法如SHAP通常不满足DAG忠实性，其扩展也有同样计算限制。

链接：https://arxiv.org/abs/2607.16236

机构：XITASO GmbH（西塔索有限公司）

作者：Amadeo Tunyi

AI 总结：研究递归推理模型量化问题，发现其因激活缩放粒度致精度崩溃，提出用逐块缩放恢复转换，应用MXInt4格式，该格式在任务中与浮点格式有竞争力，克服架构量化敏感性弱点，还能转移到ARC - AGI基准测试。

链接：https://arxiv.org/abs/2607.16237

机构：Integrated Systems Laboratory, ETH Zürich（集成系统实验室，瑞士苏黎世联邦理工学院）

作者：Thorir Mar Ingolfsson, Wajeeha Tahir, Anna Tegon, Lionnus Kesting, Gamze İslamoğlu, Luca Benini

AI 总结：研究针对人工智能评判器输出有偏差等问题，提出BACON四阶段流程，结合预算人类校准与多人工智能评判器输出。通过构建辅助特征、收集人类标签训练模型，实现总体指标估计和个体级替代评分，提高了预测准确性等，提供实用评估框架。

链接：https://arxiv.org/abs/2607.16239

机构：Adobe Research（Adobe研究院）； University of California, Berkeley（加州大学伯克利分校）

作者：Lei Shi, Anlan Zhang, Rita Lyu, Zhengmian Hu, Tong Yu, David Arbour, Avi Feller, Saayan Mitra, Ritwik Sinha

AI 总结：研究针对直接对齐算法过度优化隐式奖励模型的问题，提出添加正则化项的方法，通过研究似然变化分布理解过度优化，应用该正则化于相关方法，实现生成质量与基准能力权衡及奖励建模改进，提升了模型性能。

链接：https://arxiv.org/abs/2607.16240

作者：Shawn Im, Federico Danieli, Skyler Seto, Barry-John Theobald, Katherine Metcalf

AI 总结：研究大语言模型生成的内核是否真能超越PyTorch，指出前沿模型存在奖励黑客行为。引入KernelBench-Verified扩展评估框架及内存效率指标，实验发现最佳模型加速比低，无模型始终超PyTorch，部分模型增加GPU内存峰值使用，强调持续调整评估协议的必要性。

链接：https://arxiv.org/abs/2607.16241

机构：Meta（元（公司））； FAIR at Meta SuperIntelligence Lab（元超智能实验室公平团队）； Stanford University（斯坦福大学）

作者：Yunxiang Zhang, Ping Yu, Jianyu Wang, Max (Xiangjun)Fan, Julian Reed, Azalia Mirhoseini, Will Su

AI 总结：研究视觉语言模型在闭环交互中应对动态危险的问题，提出自进化即时记忆框架，含RSG、事实记忆和经验记忆，并通过自动测试-验证-写入循环完善元技能，实验证明该框架大幅提升安全成功率且不阻碍任务进展。

链接：https://arxiv.org/abs/2607.16247

机构：Huazhong University of Science and Technology（华中科技大学）； Tsinghua University（清华大学）； Shanghai Jiao Tong University（上海交通大学）

作者：Bingrui Sima, Lizhong Wang, Xiaoya Lu, Kun He, Xiao Yang

AI 总结：研究长上下文大语言模型推理中 KV 缓存量化问题，提出局部分布恢复技术，实现 DGAP。该技术可检测高风险步骤并恢复 top-K 候选分布，实验证明在 Llama-3.1-8B 等模型上能提升准确率、降低分布漂移，保持低比特缓存占用。

链接：https://arxiv.org/abs/2607.16248

作者：Gradwell Dzikanyanga, Yanqi Pan, Weihao Yang, Donglei Wu, Wen Xia, Hao Huang

AI 总结：研究长上下文检索中FFN的作用，通过逐层缩放模型自身FFN写入进行测试，发现其响应面带符号、特定层且任务条件化，两种单调角色可由写入梯度对齐诊断，还能预测检索衰减损伤及改进检索边界。

链接：https://arxiv.org/abs/2607.16254

机构：Ocean University of China（中国海洋大学）

作者：Zhibo Yang

AI 总结：该研究利用大语言模型解决表格数据特征生成问题，创建管道结合属性与提示生成新特征，经选择算法筛选，应用于八个不同数据集，结果显示语言模型生成的新特征有助于给定任务并提升分类效果。

链接：https://arxiv.org/abs/2607.16255

机构：University of Luxembourg（卢森堡大学）

作者：Aria Nourbakhsh, Benoît Alcaraz, Christoph Schommer

AI 总结：针对人工智能在电力电子变流器建模中面临的挑战，提出神经控制微分方程框架，用于电网形成逆变器电磁暂态仿真，通过仿射控制公式和正则化方法，实现多尺度分析，准确再现暂态响应，提供物理一致的替代建模方法。

链接：https://arxiv.org/abs/2607.16258

机构：Corporate Research Center, Midea Group（美的集团中央研究院）

作者：Jiagang Qu, Yong Tao, Dan Wang, Enyi Li, Jingjing Qi, Ding Wang

AI 总结：研究旨在解决自动化科学发现中语言模型与生物物理的认知差距。通过多尺度自主发现引擎Octopus，结合大语言模型群体和算法物理引擎，针对结直肠癌转录组进行无监督扫描，发现IGF2是5-氟尿嘧啶耐药脆弱性，建立了可验证的生物医学发现范式。

链接：https://arxiv.org/abs/2607.16262

作者：Christopher Baker, Tianyu Ren, Karen Rafferty, Hui Wang, Simon McDade

AI 总结：研究针对抗体表达排序中标记数据稀缺问题，提出基于偏好的学习框架，结合定量表达与弱监督，通过改进DPO适用于蛋白质语言模型，在多样数据集上评估，该方法优于基线，为抗体可表达性优化提供可扩展方案。

链接：https://arxiv.org/abs/2607.16263

作者：Josh Qixuan Sun, Morteza Babaie, Wenyang Hou, Mark Crowley, David Young

AI 总结：研究城市环境中路径损耗预测，提出EA-RMENet模型，采用含EfficientNetB5编码器、AG跳跃连接和ASPP的U-Net框架，平衡了准确性与效率，在相关数据集和挑战赛中有良好表现，展现出在实际无线电地图估计中的潜力。

链接：https://arxiv.org/abs/2607.16449

作者：Jonathan O'Shea (DCU School of Electronic Engineering), Conor Brennan (DCU School of Electronic Engineering)

AI 总结：研究针对现代冲突中无人机检测需求，利用轻量级卷积神经网络，将样本转换为光栅化时域图像，设计自定义模型架构并测试评估，结果显示紧凑模型能保持高精度且计算要求低，相比现有方法降低成本。

链接：https://arxiv.org/abs/2607.16455

机构：Ludovika University of Public Service（路德维卡公共服务大学）； Eötvös Loránd University（厄特沃什·罗兰大学）

作者：Gábor Farkas, Gábor Fazekas, Karakai Patrik, András Németh, Gábor Farkas

AI 总结：研究针对表格分类中过采样会扭曲特征重要性排名的问题，提出K-IPO框架，通过“生成然后选择”保留原始数据特征重要性排名，经实验评估，该方法在多方面表现出色，提升了预测性能且计算开销合理。

链接：https://arxiv.org/abs/2607.16478

机构：Department of Informatics and Telecommunications, University of Ioannina（约阿尼纳大学信息与电信系）； Industrial Systems Institute, Athena Research Center（雅典娜研究中心工业系统研究所）； Archimedes Unit, Athena Research Center（雅典娜研究中心阿基米德单位）； Department of Computer Engineering and Informatics, University of Patras（帕特雷大学计算机工程与信息学系）； Department of Mechanical Engineering and Aeronautics, University of Patras（帕特雷大学机械工程与航空系）

作者：Marios Tyrovolas, Argiris Sofotasios, Dimitris Metaxakis, Georgios Mermigkis, George Georgoulas, Panagiotis Hadjidoukas, Chrysostomos Stylios

AI 总结：研究联合故障诊断和剩余使用寿命估计的多任务深度学习模型中，滑动窗口序列划分对性能的影响。采用AMTLNet架构，引入基于块的泄漏审核划分协议，通过实验得出任务稳定性与标签来源有关，贡献了评估协议和相关证据。

链接：https://arxiv.org/abs/2607.16493

机构：Western Illinois University（西伊利诺伊大学）； Ahsanullah University of Science and Technology（阿山努拉科技大学）； Trine University（特莱恩大学）； Sapienza University of Rome（罗马第一大学）； Bangladesh University of Engineering and Technology（孟加拉工程技术大学）； Chittagong University of Engineering and Technology（吉大港工程技术大学）； Rajshahi University of Engineering & Technology (RUET)（拉杰沙希工程技术大学）

作者：Md Mahamudur Rahaman Shamim, Md. Nuruzzaman, Zannatul Ferdus, Md Rajib Ahmed, Abieer Nwshad Anward, Mohammad Tooneer, Johir Uddin Khan, Khalid Hossen

AI 总结：研究蛋白质折叠分类，将α碳接触图上的离散Ricci曲率作为轻量级结构描述符，与多种基线比较。结果显示其性能优于平均池化的ESM-2嵌入，与持久同调结合性能最强，为预训练蛋白质语言模型嵌入提供实用替代方案。

链接：https://arxiv.org/abs/2607.16553

作者：Jianru Shen

AI 总结：研究旨在提出多模态深度学习架构用于急诊分诊，利用自注意力捕捉特征关系，处理表格和文本数据。用马来西亚理科大学医院急诊科数据集验证，该模型比基线模型在准确率、F-1分数和ROC AUC上均有提升，展现出预测分诊决策的潜力。

链接：https://arxiv.org/abs/2607.16662

机构：Hospital Universiti Sains Malaysia（马来西亚理科大学医院）

作者：Hazqeel Afyq Athaillah Kamarul Aryffin, Kamarul Aryffin Baharuddin, Mohd Halim Mohd Noor

AI 总结：研究针对物理设计中可布线性估计问题，提出CLDRoute框架，将其作为条件生成问题，用物理感知条件和特定任务潜变量建模，支持样本推理，在CircuitNet 2.0上取得较好实验结果，能生成预期结果及其不确定性。

链接：https://arxiv.org/abs/2607.16674

机构：University of Connecticut（康涅狄格大学）； Tufts University（塔夫茨大学）； University of Minnesota Twin Cities（明尼苏达大学双城分校）

作者：Kiran Thorat, Nicole Meng, Caiwen Ding, Yingjie Lao, Zhijie Jerry Shi

AI 总结：研究针对海洋生物地球化学预测滞后问题，提出BG4Sea系统，通过模块化架构及多种技术如列自动编码器等，在全球海洋再分析数据上训练评估，生成多变量预测，性能超传统方法，为未来方法提供可解释基线。

链接：https://arxiv.org/abs/2607.16731

机构：Mercator Océan International（墨卡托海洋国际组织）； LIP6, Sorbonne Université（巴黎第六大学信息学实验室，索邦大学）； INRIA (ARCHES)（法国国家信息与自动化研究所（ARCHES团队））

作者：Gabriela Martinez Balbontin, Anastase Charantonis, Dominique Bereziat, Stefano Ciavatta

AI 总结：研究小语言模型中真相方向，通过无训练定向探针及多模型实验，探讨真相维度与知识的关系、架构组件作用及方向混合情况，揭示关系定律与知识门控定律，表明混合几何属知识领域。

链接：https://arxiv.org/abs/2607.16741

作者：Francesco Karim Vicidomini

AI 总结： 研究在深度生存模型中，针对小且严重删失的肿瘤学队列端到端训练时频繁计算C-index成本高的问题，提出价值单调一致性损失SCL，它与架构无关，训练时损失值与C-index强相关，能可靠替代C-index用于模型选择等。

链接：https://arxiv.org/abs/2607.16802

机构：Department of Radiation Oncology, University of Colorado School of Medicine（科罗拉多大学医学院放射肿瘤学系）； Department of Radiation Oncology, University of Texas Southwestern Medical Center（德克萨斯大学西南医学中心放射肿瘤学系）

作者：Meixu Chen, Kai Wang, Jing Wang

AI 总结：研究潜在字典学习后对活跃物理射线的推断方法，通过高斯训练-测试实验，保留兼容字典剖析测试表示，投影到支持空间，刻画相关特性，产生真实分辨率自适应支持陈述并指导训练与测试测量分配。

链接：https://arxiv.org/abs/2607.16813

机构：Institute of Data Science and Information Computing, National Chung Hsing University（国立中兴大学数据科学与信息计算研究所）

作者：Guan-Ju Peng

AI 总结：研究针对锂离子电池健康监测与设计脱节问题，提出基于虚拟传感和物理信息学习的框架，能推断设计参数，嵌入机制降低预测误差，虚拟传感减少相关误差，建立反馈回路，助力电池设计与评估。

链接：https://arxiv.org/abs/2607.16864

作者：Wendi Guo, Søren Byg Vilsen, Daniel Ioan Stroe, Yaqi Li, Yicun Huang, Ashima Verma, Daniel Brandell

AI 总结：研究实时卡车满载投标接受决策问题，基于拉格朗日松弛构建双价格策略，证明其有效性、渐近最优性及证书局限性，实验表明该策略在基准测试中表现良好，决策速度快，证书稳定。

链接：https://arxiv.org/abs/2607.16891

机构：Bubba AI（Bubba人工智能公司）

作者：Aswin Chandrasekaran

AI 总结：研究安全自适应控制中安全约束是否允许有信息实验，定义预承诺信息为承诺前学习者可见定律间KL散度，通过因果约简得出有界预承诺信息会留预言机差距，在约束线性系统中建立障碍并证明特殊情况恢复及推导相关证书。

链接：https://arxiv.org/abs/2607.16895

机构：Boston University（波士顿大学）

作者：Venkatesh Saligrama

AI 总结：研究LeJEPA的\sigreg在语言模型预训练中能否重塑表示及助于KV缓存量化，通过训练1.1亿参数模型发现，\sigreg能降隐藏状态各向异性，对K和V直接正则化可降缓存各向异性，特定配置下训练干预在量化器比例粗时有用，此为首次训练时分布正则化评估。

链接：https://arxiv.org/abs/2607.17019

机构：Auburn University（奥本大学）； National Central University（国立中央大学）

作者：Libo Sun, Po-Wei Harn, Zewei Zhang, Peixiong He, Xiao Qin

AI 总结：研究大语言模型约束推理，在接近匹配子句密度时测试实例级转移，比较不同公式和锚点，发现求解器与模型硬度差异，以及模型对证明保留重新标记敏感，还研究了完成令牌花费与代理的关系。

链接：https://arxiv.org/abs/2607.17047

作者：Lucky Verma

AI 总结：研究贝叶斯滤波变压器（BFT）在预测时内化的潜在任务先验和后验，现有预测空间比较方法脆弱。提出用预测蒙特卡罗（PMC）作为通用解释工具，通过下一个token生成近似潜在任务的隐式先验和后验，并应用于三个任务族验证，直接在潜在空间回答解释性问题。

链接：https://arxiv.org/abs/2607.17060

机构：Monash University（莫纳什大学）

作者：Afiq Abdillah Effiezal Aswadi, Haotong Ma, Susan Wei

AI 总结：研究在语言模型中学习特征时间尺度的问题，提出持久稀疏自动编码器，通过为特征学习持久性系数扩展标准SAEs，实验表明其能保持竞争力的重建质量，为解释和监测语言模型带来新机会。

链接：https://arxiv.org/abs/2607.17117

机构：University of Cambridge（剑桥大学）； University of Oxford（牛津大学）

作者：Haoyan Luo, Mateo Espinosa Zarlenga, Mateja Jamnik

AI 总结：针对阿萨姆语语音识别因数据不足面临挑战的问题，提出基于Whisper模型的可控微调系统，在特定语料库训练，并采用优化训练管道。该微调模型在多项指标上显著优于零样本基线，提升了语音识别性能。

链接：https://arxiv.org/abs/2607.17164

作者：Ganapati Das, Dwipen Laskar, Hasin Afzal Ahmed, Sanjib Kr Kalita, Kshirod Sarmah, Hem Chandra Das, Manjula Kalita

AI 总结：研究基于Transformer的语言模型在算术任务中的表现，通过分解任务、分析损失收敛等，应用人类策略和方法提升其性能，并经多种验证展示有效性，探索其与人类学习者相似性以增强关键应用中的信任。

链接：https://arxiv.org/abs/2607.17166

机构：The Hong Kong University of Science and Technology（香港科技大学）； The Hong Kong Polytechnic University（香港理工大学）； University of California, Berkeley（加州大学伯克利分校）

作者：Luyu Qiu, Jianing Li, Hwanhee Kim, Xiaoyong Wei, Yueyuan Zheng, Janet Hsiao, Lei Chen

AI 总结：研究针对纵向T细胞受体组库建模问题，提出DynImmune-BERT模型，结合多种方法，能在有纵向组库结构时补充静态编码器，通过特定评估方式得出结果，对小外部队列和协议差异需谨慎解读。

链接：https://arxiv.org/abs/2607.17244

机构：University of Macau（澳门大学）； Hanyang University（汉阳大学）； Zhejiang University（浙江大学）； Wuhan University（武汉大学）； Tsinghua University（清华大学）； South China University of Technology（华南理工大学）； University of Liverpool（利物浦大学）

作者：Rong Fu, Yongtai Liu, Xiaowen Ma, Haoyu Zhao, Shuo Yin, Yiqing Lyu, Long Zhang, Wangyu Wu

AI 总结：研究针对AIGB范式在自动出价中存在的问题，提出AIGB-R1框架，利用大语言模型推理能力，通过分层规划器与执行器模块、经验驱动循环、两阶段训练及新优化方法，经实验验证该框架在自动出价任务中的有效性。

链接：https://arxiv.org/abs/2607.17281

机构：Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）； Alibaba Group（阿里巴巴集团）

作者：Yuejia Dou, Hesong Wang, Xinyu Zhang, Tianyu Wang, Zhilin Zhang, Chuan Yu, Jian Xu, Bo Zheng, Qi Qi

AI 总结：针对计算最大间隔分离超平面问题，提出一种基于初始分离超平面的迭代几何方法，通过考虑局部信息解决凸二次优化问题，实验表明该方法在较大数据集上有竞争力，有时能超越现有直接求解方法。

链接：https://arxiv.org/abs/2607.17282

机构：Laboratory of Parallel and Distributed Systems Institute for Computer Science and Control (SZTAKI)（并行与分布式系统实验室 计算机科学与控制研究所（匈牙利科学院计算机与自动化研究所））； John von Neumann Faculty of Informatics Obuda University（欧布达大学约翰·冯·诺依曼信息学院）

作者：Akos Hajnal

AI 总结：研究神经网络验证中前瞻分支策略，提出通用集成方法，以FSB为例说明，还阐述其能生成加速验证的引理，通过在两个验证器中实例化，实现验证时间加速及解决实例数增加。

链接：https://arxiv.org/abs/2607.17290

机构：Amherst College（阿默斯特学院）； University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； Hebrew University of Jerusalem（耶路撒冷希伯来大学）； Stanford University（斯坦福大学）

作者：Liam Davis, Duo Zhou, Huan Zhang, Guy Katz, Clark Barrett, Haoze Wu

AI 总结：研究连续机器学习监测中常用漂移检测器的误报率，通过实证分析PSI、KS等五种检测器，发现不同检测器对批量大小的敏感性不同，应用邦费罗尼校正有稳定性 - 敏感性权衡，为生产ML系统选择和校准检测器提供实用指南。

链接：https://arxiv.org/abs/2607.17336

机构：Indian Institute of Technology, Roorkee（印度理工学院鲁尔基分校）

作者：Raj Shekhar Singh

AI 总结：研究针对分析布局器替代指标与下游质量不一致问题，提出基于大语言模型的CoEvoP&R框架自动进化布局目标，经实验验证该方法能有效降低线长、拥塞及时序负松弛，提升布局效果。

链接：https://arxiv.org/abs/2607.17398

机构：University of Alberta（阿尔伯塔大学）； New York University（纽约大学）

作者：Ruogu Chen, Weihua Xiao, Ramesh Karri, Jie Han

AI 总结：研究针对大语言模型在化学推理中虚构对象的问题，提出分层验证器，通过与数据库核对及门控校正减少公式错误，检索次数大幅减少，修复成功率高，但检测召回率是瓶颈，基于事实验证可提高答案质量，长尾错误处提升明显。

链接：https://arxiv.org/abs/2607.17417

作者：Can Polat, Mustafa Kurban, Erchin Serpedin, Hasan Kurban

AI 总结：研究针对专家领域树形结构与欧几里得变换器的问题，提出HySAT方法，即仅在损失层使用双曲损失。通过构建和部署六个专家语言模型进行实验，证明该方法的有效性及稳定性，四个模型已实际应用，有相关记录。

链接：https://arxiv.org/abs/2607.17513

机构：PolymathMinds Lab（多智思维实验室）； aSSIST University（aSSIST大学）； Samsung Engineering（三星工程公司）

作者：Kwan Soo Shin, In Seok Kang, Munho Lee

AI 总结：受掩码离散扩散启发，研究高斯随机场模型单步选择，用局部相关非负分数场表示不确定性，通过距离相关高斯模型衡量所选位置相关性，建立两个互补结果，为理解相关因素对掩码离散扩散中基于置信度一步选择的影响提供基线。

链接：https://arxiv.org/abs/2607.17522

机构：University of Pennsylvania（宾夕法尼亚大学）

作者：Linjun Li

AI 总结：研究多供应商大语言模型服务基础设施故障模式，提出FailureAtlas双轴分类法，依起源层和可检测性分类，通过公共报告和测试填充，发现最严重故障是隐蔽的，如并发竞争致历史丢失和流索引冲突，还给出重现脚本。

链接：https://arxiv.org/abs/2607.17525

机构：Metriqual（梅特里夸尔）

作者：Vishal Pandey, Gopal Singh

AI 总结：研究提出结合大语言模型与神经仿真推理的框架，用于联合模型选择与参数估计。给定自然语言描述，大语言模型提出候选程序，经反馈驱动变异和神经密度估计评估，能在一组模型上推理，在多基准测试中可从提示识别合理模型族。

链接：https://arxiv.org/abs/2607.17540

机构：Anthropic

作者：Siddharth Mishra-Sharma

AI 总结：针对高频金融市场极端价格变动检测难题，提出基于波动率感知的方法，利用比特币限价订单簿数据，通过扩展目标公式、采用XGBoost模型及相关验证评估，显著提升检测罕见事件能力，凸显目标设计在金融机器学习中的关键作用。

链接：https://arxiv.org/abs/2607.17555

作者：Maorufa Zaman, Haris Md Sahed

AI 总结：研究针对大语言模型结构化剪枝中独立评分单元不合理的问题，提出CoCurve方法，通过二阶泰勒展开得到Fisher矩阵计算协同剪枝曲率边，联合修剪注意力和FFN单元，仅用M次前向传播求解预算二次规划，无需标签、微调等完成剪枝。

链接：https://arxiv.org/abs/2607.17568

机构：College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算与数据科学学院）； Interdisciplinary Graduate Programme, Nanyang Technological University（南洋理工大学跨学科研究生项目）； School of Electrical and Electronic Engineering, Nanyang Technological University（南洋理工大学电气与电子工程学院）

作者：Zhiren Gong, Zihao Zeng, Zijie Wang, Tiantong Wang, Chau Yuen, Wei Yang Bryan Lim

AI 总结：针对现代深度学习中近似最近邻搜索难以兼顾功能与性能的问题，提出ANNLib框架，基于图算法解耦并优化算法和数据结构组件，集成多种算法和数据结构，为应用提供简单接口且性能更优。

链接：https://arxiv.org/abs/2607.17582

机构：UC Riverside（加州大学河滨分校）； W&M（威廉与玛丽学院）

作者：Zheqi Shen, Jingbo Su, Zijin Wan, Yan Gu, Yihan Sun

AI 总结：研究从观察到的细胞水平扰动特征中检索靶点和化合物的问题，提出Transformer检索模型\model，通过联合训练编码器映射特征，在Tahoe-100M条件下评估，结果表明该模型能从细胞反应中恢复靶点注释和化合物身份，但对未见情况的泛化待确定。

链接：https://arxiv.org/abs/2607.17671

机构：AIFFEL Research, Modulabs（艾菲研究公司，模块实验室）

作者：Kseniia Vaniushkina, Jeongmin Lim, Jinyong Park

AI 总结：研究大语言模型解决规划问题的不足，提出计算链架构，利用结构化上下文窗口，让语言模型学习规划策略、预测世界模型并执行算术运算，在多个任务上取得高成功率，能解决复杂汉诺塔问题且减少训练数据。

链接：https://arxiv.org/abs/2607.17710

机构：Department of Computing Science, University of Alberta（阿尔伯塔大学计算科学系）； Alberta Machine Intelligence Institute (Amii)（阿尔伯塔机器智能研究所）

作者：Ehsan Futuhi, Nathan R. Sturtevant

AI 总结：研究针对大语言模型推理中4位量化因异常值致精度降的问题，提出MXSens方法，基于列和层灵敏度分配混合尾数比特宽度，无需训练，利用MXINT块结构，在多模型任务中优于现有方法，平衡了量化的准确性与资源效率。

链接：https://arxiv.org/abs/2607.17733

机构：EPFL（洛桑联邦理工学院）； Tsinghua University（清华大学）； MangoBoost Inc.（芒果助推公司）； Korea University（韩国大学）； Google DeepMind（谷歌深度思维）

作者：Simla Burcu Harma, Danila Mishin, Zhengyuan Su, Ayan Chakraborty, Elizaveta Kostenok, Dongho Ha, Babak Falsafi, Martin Jaggi, Yunho Oh, Amir Yazdanbakhsh

AI 总结：研究预测过程监控中深度学习模型黑箱问题，提出基于控制流感知分割算法的局部事后可解释性方法，能计算段级SHAP解释，在合成及真实数据集上评估验证，提升模型可解释性。

链接：https://arxiv.org/abs/2607.17783

机构：SKEMA Business School（SKEMA商学院）； KU Leuven（鲁汶大学）

作者：Kseniya Sahatova, Rafael Seidi Oyamada, Xuefei Lu, Johannes De Smedt

AI 总结：探讨机器学习中表征概念，随着模型发展其从工程领域转向心理表征领域。评估《柏拉图式表征假说》中关于不同AI模型表征属性趋同由现实统一结构驱动的说法，借助心灵哲学观点审视，阐明关键、解释不足并给出研究方向。

链接：https://arxiv.org/abs/2607.17800

作者：Gilad Landau, Aviv Keren

AI 总结：研究挑战Transformer中块作用与位置绑定的传统观点，提出基于循环深度折叠的莫比乌斯学习，同一组块在不同数据流中有不同应用时机，实现深度角色叠加，实验表明其在分布式训练中有优势，开辟新设计空间。

链接：https://arxiv.org/abs/2607.17843

机构：Zhejiang University（浙江大学）

作者：Tongtian Zhu

AI 总结：研究提出CMP架构，通过局部、无梯度更新学习，测试其抵御灾难性遗忘能力。在文本域实验中，CMP反向转移比Transformer好15 - 19倍，还报告了与基线差距等结果，表明该架构在特定条件下抗灾难性遗忘效果优于反向传播。

链接：https://arxiv.org/abs/2607.17944

机构：Arkadhi Labs（阿卡迪实验室）

作者：Ashmith Atmuri, Akshay Kumar, Yashaswini Rao Bhogarajula

AI 总结：在MLSys 2026 FlashInfer竞赛中，针对NVIDIA Blackwell B200 GPU，提出以工具为中心的大语言模型驱动的GPU内核优化系统，分离评估工具与优化控制器，利用Codex等生成候选内核，实验显示优化后平均延迟加速显著，且代理辅助内核效果更佳。

链接：https://arxiv.org/abs/2607.17979

机构：Baidu, Inc.（百度公司）

作者：Yue Shui, Chenyu Ma, Hangfei Xu, Shengzhao Wen, Yanpeng Wang

AI 总结：研究针对缩放点积注意力在Transformer模型中作为相似性度量的局限，提出L1增强注意力方法，通过减去特定头的L1距离改进相似性计算，经实验在WikiText 2上取得更好效果，揭示了各层几何作用及头级专业化，有效提升语言模型相似性计算。

链接：https://arxiv.org/abs/2607.18027

作者：Kurt Godden

AI 总结：针对大语言模型在边缘设备部署的挑战，提出SelectInfer框架，通过离线分析识别特定任务和通用神经元，实施选择性加载和计算两项关键优化，可显著减少内存与计算量，同时保持任务性能，推动LLM在边缘设备的部署。

链接：https://arxiv.org/abs/2607.18081

机构：Paderborn University（帕德博恩大学）

作者：Huzaifa Shaaban Kabakibo, Eric Schniedermeyer, Artem Burchanow, Lin Wang

AI 总结：研究针对资源受限硬件上设备端模型自适应难题，提出用异构自适应管道，借助Hailo-8L加速器提取冻结主干特征，划分计算图，在多架构和数据集上提升训练速度、吞吐量并降低能耗，证明该方法实用。

链接：https://arxiv.org/abs/2607.18101

机构：Poznan University of Technology（波兹南理工大学）； University of Modena and Reggio Emilia（摩德纳大学和雷焦艾米利亚大学）

作者：Mateusz Piechocki, Alessandro Capotondi, Marek Kraft

AI 总结：研究实时多模态应用部署难题，提出FlashRT智能体框架。通过新范式引导编码智能体多阶段转换，将参考实现转为高效部署，在不同GPU上显著提升性能，尤其在专家优化不成熟平台更具扩展性。

链接：https://arxiv.org/abs/2607.18171

机构：Carnegie Mellon University（卡内基梅隆大学）； AMD（超威半导体公司）； University at Buffalo（纽约州立大学水牛城分校）

作者：Krish Agarwal, Zhuoming Chen, Yanyuan Qin, Zhenyu Gu, Atri Rudra, Beidi Chen