机器学习学术速递[7.15]

AI 总结：研究Transformer中FFN模块内部结构难解释问题，引入无需训练的归因方法，发现其神经元层面存在稀疏且结构化的层间依赖关系，该方法为电路级可解释性提供工具并识别候选稀疏路径。

链接：https://arxiv.org/abs/2607.11990

作者：Johannes Knittel, Hanspeter Pfister

英文摘要：Feedforward network (FFN) blocks account for a large fraction of the parameters and computation in Transformer architectures, yet their internal structure remains difficult to interpret due to the additive superposition induced by the residual stream. We examine whether the activation of an FFN neuron can be explained by a sparse set of preceding neuron activations and attention outputs. We introduce a training-free attribution method that estimates the relative influence of upstream neurons and attention outputs on a target neuron's activation. Empirically, across models and layers, we find that small subsets of preceding activations and attention outputs suffice to preserve neuron activations with high fidelity when all remaining inputs are masked with their average values. Effective sparsity is even greater when accounting for the inherent activation sparsity of upstream layers. Moreover, applying the neuron-specific masks in all layers simultaneously, such that the induced deviations propagate through the network, leaves model perplexity largely unchanged at moderate sparsity levels. These results demonstrate that, despite dense parameterization, FFNs exhibit sparse and structured inter-layer dependencies at the neuron level. Our method provides a practical, scalable tool for circuit-level interpretability and identifies candidate sparse pathways with potential implications for efficient inference.

AI 总结：研究通过交互感知专家混合模型，利用结构化健康记录多层次视图预测中风后僵硬，路由归因揭示视图差异，凸显视图构建对可解释性的关键作用。

链接：https://arxiv.org/abs/2607.12255

作者：Ji Hwan Park, Ying Ding, Tianjin Guo

英文摘要：We study interaction-aware mixture-of-experts for post-stroke rigidity prediction using multi-level views of structured health records. Despite minimal performance gains, routing attribution reveals systematic importance differences across views, underscoring view construction as key to interpretability.

AI 总结：研究如何解释神经网络内部特征从特定输入中提取的信息，提出源接地特征反演方法，通过特定条件和映射修复信号，经DAG反向传递组合状态，支持多种架构，验证反演依赖关系，打开模型隐藏特征层次结构。

链接：https://arxiv.org/abs/2607.12526

作者：Kaixiang Shu

英文摘要：Interpreting a neural network requires understanding what its internal features extract from a particular input. Feature inversion seeks to express a selected feature in the input domain, but canonical iterative methods search for an input whose re-encoded representation matches the target. Because many inputs can satisfy this constraint, target matching alone does not specify the inverse associated with the sample that generated the feature. We formulate source-grounded feature inversion by conditioning the inverse on the source-local network geometry at the target-generating input. At each boundary of the computational DAG, backpropagation provides the correct reverse dependencies but transports an adjoint signal rather than an upstream-state estimate. We locally repair this signal with a closed-form matrix Wiener map from a mean-seed VJP to the upstream state, followed by a second Wiener map for the JVP forward-consistency residual, and compose the repaired states through the same DAG in one finite reverse pass. One calibrated zero-intercept map family supports new inputs, depths, channels, and channel groups across diverse CNN and Transformer architectures, tensor components, and visual distributions without query-specific optimisation. Matched target and source controls verify that each inverse depends on the selected feature and the local operators of the sample being explained, rather than a target-independent image template. Prediction-conditioned feature atlases align these visualisations with independent interventions on the corresponding internal features. Together, source-grounded feature inversion opens the model's hidden feature hierarchy to inspection at the level of individual layers and channels, linking what the network extracts from an input to the internal evidence that shapes its decision.

AI 总结：本书研究数据科学的数学基础，涵盖高维特性、多种分析方法如奇异值分解、线性回归等，以及图、网络、聚类等内容，为数据科学提供了全面的数学理论支撑。

链接：https://arxiv.org/abs/2607.11938

作者：Afonso S. Bandeira, Amit Singer, Thomas Strohmer

英文摘要：This book is about the mathematical foundations of data science. 1. Introduction 2. Curses, Blessings, and Surprises in High Dimensions 3. Singular Value Decomposition and Principal Component Analysis 4. Linear Regression and Regularization 5. Graphs, Networks, and Clustering 6. Nonlinear Dimension Reduction and Diffusion Maps 7. Linear Dimension Reduction via Random Projections 8. Optimization for Data Science 9. Classification 10. A Mathematical Introduction to Deep Learning 11. Large Sample Limit of Graph Laplacians 12. Community 13. Concentration of Measure and Gaussian Analysis 14. Matrix Concentration Inequalities 15. Compressive Sensing and Sparsity 16. Low-Rank Matrix Recovery

AI 总结：研究针对自动驾驶中多源观测冗余问题，提出LIDAR-AD，用减少冗余的潜在对齐取代观测重建，将车辆控制建模为残差动作更新，经实验验证其在模拟场景和现实布局下性能优异，能提升风险感知等能力。

链接：https://arxiv.org/abs/2607.11964

机构：School of Mechanical Engineering, Southeast University（东南大学机械工程学院）； Department of Civil and Environmental Engineering, University of Michigan（密歇根大学土木与环境工程系）； Xheart Technology Co., Ltd.（芯驰科技有限公司）

作者：Yongzhi Liu, Yang Xiao, Zhong Cao, Zeng Kang, Sunan Zhang, Zhaozhi Dong, Guojun Yu, Weichao Zhuang

英文摘要：Autonomous driving requires long-horizon closedloop decision making in dynamic traffic environments. Latent world models offer an effective framework for this problem by enabling imagination-based decision making in compact latent spaces. However, multi-source observations contain controlirrelevant redundancy, whereas reliable driving decisions rely on risk-relevant relations, future dynamics, and continuous action adjustments. This mismatch makes observation reconstruction and absolute action modeling suboptimal for learning decisionrelevant latent dynamics. We propose LIDAR-AD, a decoderfree Latent-Interaction Dreamer with Action-Residual Chains for autonomous driving. LIDAR-AD replaces observation reconstruction with redundancy-reduced latent alignment, encouraging compact representations of risk-relevant relations in multi-source driving inputs. It further models vehicle control as residual action updates and uses residual-action sequence contrastive learning to align multi-step residual-driven rollouts with future latent states. A deterministic analysis shows that the latent-tanh residual parameterization preserves interior action reachability while representing smooth long-horizon control as compact local updates. Together, these designs improve risk-aware state abstraction, continuous-control modeling, and long-horizon dynamics prediction. Extensive experiments across diverse simulated driving scenarios demonstrate that LIDAR-AD consistently outperforms world-model baselines, achieving the highest reward and the best success rate among learning-based methods. Evaluations on nuPlan-derived log-reconstructed scenarios further demonstrate the transferability of LIDAR-AD under real-world traffic layouts.

AI 总结：研究提出能否让替代模型在预测场演化前学习分辨率位置的问题，构建两阶段扩散框架解决自适应离散化，通过多种约束使网格生成器正则化，结果显示该方法在五个PDE regime有竞争力，表明离散化应依 regime学习，重构了自适应网格划分问题。

链接：https://arxiv.org/abs/2607.11974

作者：Zixuan Shen (1), Bingchuan Wang (1), Zhi Wang (2), Yong Wang (1) ((1) Central South University, (2) Nanjing University)

英文摘要：Most neural partial differential equation (PDE) surrogates learn how fields evolve after a grid has already been chosen. However, before any operator is applied, the grid has already determined how modeling capacity is allocated across space, resolution, and spectral bandwidth. We argue that this hidden design choice should itself be learnable, leading to a question different from standard operator learning: can a surrogate learn where resolution should exist before predicting field evolution? We formulate adaptive discretization as a physics-constrained conditional generation problem over valid mesh displacements. The success of diffusion models in PDE field prediction suggests their potential for learning adaptive discretizations under similar structured constraints. This leads to a two-stage diffusion framework: Stage 1 learns an r-adaptive displacement mesh conditioned on the observed dynamics, while Stage 2 predicts the solution evolution from the mesh-informed representation. The mesh generator is regularized by physics-aware proxy channels, geometric validity constraints, and local spectral concentration so that adaptation remains physically interpretable and numerically legal. Across five PDE regimes, the results show that diffusion-based learned discretization is competitive with adaptive-mesh and reduced-order baselines, with particularly strong gains in regimes where fixed or handcrafted allocation is insufficient. The main conclusion is not that there exists a universal optimal mesh rule, but that discretization should be learned in a regime-dependent manner: different spatial and spectral structures favor different allocation behaviors. This reframes adaptive meshing for neural PDE solvers from a solver-specific heuristic into a generative representation-learning problem.

AI 总结：针对网络安全漏洞分类中类别不平衡及层次依赖问题挑战，提出层次感知RoBERTa框架，通过可学习父类嵌入纳入CWE结构信息，实验证明其优于过采样方法，在CWE研究概念数据集上加权F1分数达0.76，少数类提升显著。

链接：https://arxiv.org/abs/2607.11994

机构：Texas Tech University（德克萨斯理工大学）； Cumberland University（坎伯兰大学）； Yeshiva University（叶史瓦大学）； University of Cumberlands（坎伯兰大学）

作者：Bipin Chhetri, Deepika Giri, Avishek Kadel, Rabin Kumar Karki, Akbar Siami Namin

英文摘要： Classifying cybersecurity vulnerabilities using the Common Weakness Enumeration (CWE) taxonomy is challenging due to extreme class imbalance and strong hierarchical dependencies among weakness categories. Although oversampling techniques such as Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN) are widely adopted to mitigate class imbalance, their effectiveness for hierarchical CWE text classification remains largely unexplored. This paper proposes a Hierarchy-Aware RoBERTa framework that explicitly incorporates CWE structural information through learnable parent-class embeddings, preserving taxonomic consistency. Our experiments demonstrate that synthetic interpolation in high-dimensional embedding spaces violates the inherent parent-child constraints of the CWE hierarchy, offering only marginal benefits for classical ML models while consistently degrading deep learning architectures. Evaluated on a CWE Research Concept dataset, the proposed model achieves a weighted F1-score of 0.76 without data augmentation, outperforming all baselines with notable gains on minority classes, including the Class category whose F1-score improved from 0.40 to 0.60 over the BERT baseline. Our results suggest that hierarchy-aware representation learning is a more principled alternative to oversampling for structured vulnerability classification.

AI 总结：该研究针对药物发现难题，提出基于条件扩散的SBDD框架conDitar-dev，含msPRL、conDitar和paOPT三个模块。在新基准上表现出色，在ADMET属性上性能提升显著，应用于两个靶点验证了其生成可开发分子的能力，还发现了新抑制剂及药物重新定位机会。

链接：https://arxiv.org/abs/2607.12349

机构：The Ohio State University（俄亥俄州立大学）

作者：Ruoxi Gao, Jiangweizhi Peng, Ziqi Chen, Frazier N. Baker, David C. Kombo, John L. Kane Jr., Andrew A. Scholte, Yi Li, Matthew J. LaMarche, Luigi I. Iconaru, Hans-Peter Biemann, Mingyi Hong, Xia Ning

英文摘要：Drug discovery and development is time-consuming and resource-intensive, motivating computational approaches such as diffusion models for de novo drug design. Many such models follow the structure-based drug design (SBDD) paradigm, generating molecules to fit a target binding pocket. However, existing diffusion-based SBDD methods typically couple pocket and ligand representation learning, model interactions only at the atom level, and prioritize binding affinity over other developability properties. Here, we introduce conDitar-dev, a conditional diffusion-based SBDD framework for generating ligands with strong binding affinities and favorable ADMET properties. It consists of three modules: msPRL, a pretrained multi-scale pocket representation learning module; conDitar, a pocket-conditioned diffusion model guided by msPRL representations; and paOPT, a generation-time method for optimizing ligand developability. On a newly curated benchmark of human disease targets, conDitar outperforms state-of-the-art SBDD baselines, achieving an average binding score of -8.85 kcal/mol. Across five ADMET properties, conDitar-dev improves performance by up to 73% over conDitar. To further validate the abilities of conDitar-dev to generate developable molecules, we have applied it to two validated druggable targets: programmed death-ligand 1 (PD-L1) and colony-stimulating factor 1 receptor (CSF1R) proteins. Top-ranked generatively designed molecules and their analogs have been experimentally synthesized and biologically tested. Two molecules generated directly by conDitar-dev for PD-L1 exhibited SPR-derived $K_D$ values of 3.49 and 3.75 $\mu$M, respectively. Hit expansion based on conDitar-dev-designed molecules identified selective CSF1R inhibitors with IC$_{50}$ values as low as 200 nM, while also uncovering opportunities for drug repositioning.

AI 总结：研究强化学习中高奖励与潜在状态理解的关系，通过将任务表示为隐藏自动机，利用三个可控轴（优化器强度、任务结构、观察信息量）来分别测量奖励成功和潜在状态学习，区分感知差距和规划差距，得出高奖励非任务理解证据且智能体恢复潜在状态可预测的结论。

链接：https://arxiv.org/abs/2607.11953

作者：Jim Allchin

英文摘要：Does a reinforcement-learning agent that earns high reward represent its task's latent state, or only a reward-correlated shortcut? The question is usually unanswerable: the "true state" is undefined. We make it exactly answerable with a white-box instrument: express the task as a hidden deterministic finite automaton (DFA), let the agent observe a symbol stream and intermittently choose the next symbol under partial control, and grant one sparse terminal reward for acceptance. Knowing the automaton gives two things for free: the optimal return (so reward becomes an interpretable normalized score) and the exact latent state at every step (so we can probe the agent's representation without ever showing it). Reward success and latent-state learning thus become separately measured quantities whose coupling is governed by three controllable axes. Optimizer strength: under weak on-policy RL the agent earns reward with the state probe at chance for every architecture, tempting the conclusion that sparse RL cannot install latent state; a pre-registered control overturns it -- PPO+GAE recovers the state, but only partially and with high seed variance. Task structure: permutation (group-language) structure is a warning sign computable from the transition function before any training, and held out on 153 capacity-controlled fresh automata it flags perception gaps at precision 0.86 (89 of 103), in one direction only. Observation informativeness: a label-free auxiliary is vacuous when observations carry no state and recovers it in proportion to how much they reveal. The payoff is a distinction reward-only evaluation cannot make: a perception gap (latent state not linearly recoverable, though representable) versus a planning gap (state recoverable but unused). High reward is thus not evidence of task understanding; whether an agent recovers latent state is predictable in advance.

AI 总结：研究针对临床代码预测任务，提出图约束遍历策略，将其转换为有限时域决策过程。单一语言模型逐层选节点得叶代码，实现极端多标签预测到子集决策的转换。实验表明该策略优于平面基线，增加监督数据可提升性能，简单图约束策略学习效果良好。

链接：https://arxiv.org/abs/2607.11954

作者：Amritpal Singh, Sebastian Torres, Khawar Shakeel, Syed Ahmad Chan Bukhari

英文摘要： Clinical code prediction maps unstructured discharge summaries to ICD-10-CM leaf codes in a large, sparse, and deeply hierarchical label space. Most systems treat the task as flat multi-label classification, scoring codes independently and providing limited training signal for rare labels. We propose a graph-constrained traversal policy that formulates ICD prediction as a finite-horizon decision process over a pruned code hierarchy. A single language model descends the graph level by level, selecting valid child nodes until billable leaf codes are reached. This converts extreme multi-label prediction into sparse, hierarchy-aware subset decisions while guaranteeing structurally valid outputs. On MIMIC-IV discharge summaries, our best supervised policy, SFT-1+, achieves 0.709 micro-F1 on a curated 50-code subset and 0.527 micro-F1 on the full 15,761-code space, outperforming flat baselines including CAML, LAAT, and PLM-ICD. In the full setting, SFT-1+ improves over the strongest flat baseline by 0.044 micro-F1 and 0.157 macro-F1, suggesting that graph-constrained decomposition mitigates the rare-code bottleneck. A controlled factorial study evaluates architecture, training algorithm, and data budget. Across both scales, one shared policy matches a three-specialist cascade while avoiding its context-window overflow on 28-32% of full-space test notes. Increasing supervised trajectory data is the only intervention that consistently improves performance, while GRPO reinforcement learning provides no benefit over supervised continuation with matched data. These results show that simple graph-constrained policy learning can outperform more complex flat, cascaded, and reinforcement-learning alternatives for extreme clinical code prediction.

AI 总结：研究微电网能量协调中联邦强化学习的约束感知聚合，提出含本地性能和约束违反的聚合规则，基于惩罚的规则效果最佳，实验表明轻量级聚合策略能在保留通信协议时提高联邦强化学习的经验安全性。

链接：https://arxiv.org/abs/2607.12763

机构：School of Computer Science, University of Galway（戈尔韦大学计算机科学学院）

作者：Usman Haider, Karl Mason

英文摘要：Federated Reinforcement Learning (FedRL) enables coordination of distributed energy resources without sharing raw local data, but standard aggregation methods such as FedAvg do not account for system-level constraints, often leading to unsafe global behavior. In this work, we study constraint-aware aggregation for federated reinforcement learning in distributed energy coordination. We propose aggregation rules that incorporate both local performance and estimated constraint violation into the server-side update. Among these, a simple penalty-based rule, $w_i \propto R_i - \alpha V_i$, consistently provides the most reliable trade-off between reward and safety, without requiring dual optimization or modifications to local training. \textcolor{black}{We evaluate our approach on DairyGridEnv, a benchmark modeling multiple farms coordinating battery storage under stochastic demand and a shared grid capacity constraint, and further assess robustness using real load-driven demand profiles from Finland and the German FIELD dataset. Across multiple seeds, penalty-based aggregation substantially reduces violations while improving reward relative to FedAvg in both synthetic and real load-driven settings.} A combined reward-violation scheme exposes a tunable trade-off via $\lambda$, but is less stable. These results demonstrate that lightweight aggregation strategies can substantially improve empirical safety in federated reinforcement learning while preserving standard communication protocols.

AI 总结：研究针对建筑物热能存储控制中模型难以扩展的问题，采用基于可验证奖励的强化学习，通过强化微调训练开放权重推理模型，在简单办公楼TES基准测试中取得较好效果，明确了规划模式转移情况，推动了相关测试与验证器发展。

链接：https://arxiv.org/abs/2607.12856

机构：The University of Tokyo（东京大学）； Tokyo University of Science（东京理科大学）

作者：Takumi Shioda, Kohei Terashima, Tatsuo Nagai

英文摘要：Buildings are expected to shift cooling loads in response to grid conditions. Thermal energy storage (TES) enables this shift, but scheduling it well requires planning hours ahead under storage constraints. Model predictive control (MPC) and reinforcement learning are difficult to scale across buildings. This study instead adapts an open-weight reasoning model through reinforcement learning with verifiable rewards (RLVR). We convert exact offline dynamic-programming (DP) action values into dense rewards for every candidate action. Using only 30 training prompts, reinforcement fine-tuning (RFT) trains the model as an upper-level scheduler that outputs hourly heat-pump setpoints from text-based states and forecasts. Evaluation uses a deliberately simple office-building TES benchmark where exact DP is tractable and the optimum is known. RFT reduces the open-weight model's emissions from 70.5 to 61.2 kg-CO2, close to the DP optimum of 60.8 kg-CO2. GPT-5 nearly matches DP and MPC without task-specific training, while GPT-4o, a non-reasoning LLM, produces higher emissions than the no-storage baseline, so inference-time reasoning appears important. Trace analysis shows that RFT mainly stabilizes observable planning patterns (candidate comparison, look-ahead, and feasibility checking) rather than creating a new strategy. Robustness and generalization tests clarify what transfers: the reinforced planning patterns persist under forecast errors and an unseen TES condition and carry over to a battery task, but its different structure limits the gains. DP-based verifiable rewards offer a practical way to adapt open-weight reasoning models to building storage scheduling. These results motivate higher-fidelity tests of whole-building control and scalable verifiers for city-scale energy management.

AI 总结：研究旨在训练强大自动驾驶智能体，提出TerraZero过程式驾驶模拟器和自博弈训练堆栈，其速度快、逼真且多样。通过特定方式填充地图生成无限场景，智能体仅靠强化学习训练，能零样本泛化，在多个基准测试中表现出色。

链接：https://arxiv.org/abs/2607.13028

机构：Applied Intuition（应用直觉）

作者：Zhouchonghao Wu, Akshay Rangesh, Weixin Li, Wei-Jer Chang, Zachary Lee, Tim Wang, Wei Zhan

英文摘要： Training robust autonomous driving agents requires a simulator that is fast enough for reinforcement learning at scale, realistic enough to ground behavior in real-world map structure, and diverse enough to cover the safety-critical long tail that logged data rarely contains. We present TerraZero, a procedural driving simulator and self-play training stack. A configurable C engine runs simulation on the CPU and policy inference on the GPU over a zero-copy path, sustaining 1.3M agent-steps per second on a single server-grade GPU, far faster than existing object-level simulators, while keeping fidelity lighter single-agent systems omit: heterogeneous agents, multiple dynamics models, and full traffic-rule enforcement. TerraZero treats logged data only as a source of real-world map geometry, populating each map with randomized rule-based road users and signal controllers and randomizing agent dynamics, rewards, and sizes per episode, so a map yields an unbounded set of scenarios. Every reported policy trains from scratch by reinforcement learning alone on a compute-efficient self-play recipe across GPUs, with zero human demonstrations and no fallback planner at inference. Policies generalize zero-shot across cities and datasets, including emergent left-hand-traffic driving without explicit supervision. As an ego policy, TerraZero is the first fully learned policy to top the InterPlan long-tail benchmark, ahead of larger learned planners; on routine-driving val14 it ranks among the best approaches and is the safest, posting the best collision and time-to-collision scores. On Waymo Open Sim Agents realism the same recipe outperforms other demonstration-free methods and is competitive with the strongest reference-anchored self-play method. One stack serves both roles: driving policies across dynamics for cars and trucks, and sim agents that jointly control vehicles, pedestrians, and cyclists.

AI 总结：研究在自由能原理下能否使预测系统发生本体反转，通过卷积变分自编码器与循环潜在预测器配对的代理进行研究，发现表征能力与默认行为解耦，存在认知复发现象，证明了抵抗现实采用是一种结构属性。

链接：https://arxiv.org/abs/2607.11958

作者：MD Ibrahim Hossain Ridoy

英文摘要：Under the free energy principle, a predictive system does not observe reality directly; it maintains a generative model of the world and experiences that model's best current hypothesis. Can a synthetic environment be made consistent enough that a predictive system's own inference machinery adopts it as this default hypothesis, permanently displacing the environment that first shaped it? We call this state ontological inversion. Because inducing and monitoring such a transition in a nervous system is neither ethical nor technically feasible, we study the underlying computational problem through a controlled proxy: a convolutional variational autoencoder paired with a recurrent latent predictor, whose evidence lower bound objective is mathematically identical, up to sign, to variational free energy itself. The network is trained first on a baseline visual domain, then on a mixed stream in which a swept rehearsal ratio r controls how much baseline content persists during transition to a target domain. Representational capacity, what the latent space can discriminate, is tracked separately from default behavior, what the system generates when left unconstrained. Across a full sweep of 90 runs, the two diverge sharply: representational accuracy stays near ceiling, 0.97 to 0.998, regardless of r, while default behavior spans nearly the system's entire range depending on r alone, a decoupling of learning from acceptance. More strikingly, at intermediate r the system's default output rises toward the target domain, then partially reverts toward the baseline while training continues unchanged, a structural failure we term cognitive relapse. Resistance to reality-adoption is not reducible to learning speed; it is a structural property with its own distinct failure modes, established here as a computational existence proof and nothing further.

AI 总结：研究针对小分子、晶体和蛋白质跨域生成管道分散问题，提出单架构流匹配自动编码器SinAE，用普通Transformer编码器和解码器，将重建负担转移到迭代流匹配解码器，实现跨域近无损重建，在生成基准测试中表现出色。

链接：https://arxiv.org/abs/2607.12380

机构：National University of Singapore（新加坡国立大学）； Tencent AI for Life Science Lab（腾讯生命科学人工智能实验室）

作者：Yuxuan Ren, Fan Yang, Jianhua Yao, Yatao Bian

英文摘要：Small molecules, crystals, and proteins all reduce to atoms in 3D space, yet their generative pipelines remain fragmented across domains, each with its Small molecules, crystals, and proteins all reduce to atoms in 3D space, yet their generative pipelines remain fragmented across domains, each with its own graph, equivariant, or frame-based architecture. Cross-domain training would mitigate per-domain data scarcity, but direct generation in 3D coordinate space cannot easily handle the heterogeneous structural priors of all three domains, and no prior latent autoencoder is simultaneously lossless and architecturally general across all three. We introduce SinAE, a single-architecture flow-matching autoencoder for molecules, crystals, and proteins, with vanilla Transformer encoder and decoder and no equivariant, graph, or domain-specific operators. Rather than requiring the encoder to capture fine-grained geometry, SinAE shifts the reconstruction burden into an iterative flow-matching decoder, achieving near-lossless reconstruction across domains and reducing reconstruction errors by orders of magnitude relative to prior latent baselines. The same per-token latent supports a standard Diffusion Transformer prior that reaches strong performance on molecular, crystal, and protein generation benchmarks. Joint molecule--crystal training strictly improves both domains, providing direct evidence of cross-domain transfer through a shared atomic latent. Code is available at this https URL.

AI 总结：研究如何让智能体从图像序列构建世界结构化地图。核心方法是将CSCG重构成可微模块gradCSCG并与VQ-VAE前端耦合，有软发射前向传播和损失平衡机制。主要贡献是证明该方法能在多环境下从视觉输入高精度恢复邻接图，CSCG可作深度学习构建块。

链接：https://arxiv.org/abs/2607.12382

机构：Goethe University Frankfurt（歌德大学法兰克福分校）； Max Planck Institute for Human Development（马克斯·普朗克人类发展研究所）； Max Planck Institute for Empirical Aesthetics（马克斯·普朗克实验美学研究所）； SINTEF（挪威科技工业研究院）； University Medical Center Göttingen（哥廷根大学医学中心）； Institute of Computer Science and Campus Institute Data Science, University Göttingen（哥廷根大学计算机科学与数据科学研究所）； Bridging Research in AI and Neuroscience (brAIN), Computer Vision Center（人工智能与神经科学交叉研究中心（brAIN），计算机视觉中心）； Computer Science Department, Universitat Autònoma de Barcelona（巴塞罗那自治大学计算机科学系）； Epilepsy Center Frankfurt Rhine-Main, Department of Neurology, Goethe University Frankfurt（歌德大学法兰克福分校莱茵-美因癫痫中心，神经学系）； Circulant Labs（循环实验室）

作者：Arash Nikzad, Sasan Sarbishegi, Ali Dasmeh, Muhammad Asif, Parsa Gharavi, Erik Husom, Sagar Sen, Andrew B. Lehr, Olivier Penacchio, Ana Clemente, Tristan M. Stöber

英文摘要： How can an agent build a structured map of its world from nothing but an ongoing sequence of raw sensory input and its own movements, especially when natural variation means exact sensory patterns rarely repeat? The Clone-Structured Causal Graph algorithm (CSCG), a normative hippocampus model, shows how an interpretable map can be learned from aliased observations. However, CSCG requires a predefined discrete alphabet, and its expectation-maximization formulation is not easily combined with existing neural network modules, preventing the end-to-end processing of raw image sequences. We remove this barrier by reformulating CSCG as a single, fully differentiable module, gradCSCG, and coupling it to a learned vector-quantized variational autoencoder (VQ-VAE) perceptual front-end. A soft emission forward pass allows the map-learning objective to flow back into perception, while a set of loss-balancing mechanisms mitigates module collapse during joint training. We demonstrate, first, that gradient training reproduces CSCG's results on original symbolic grid worlds by recovering room topology from heavily aliased observations. Second, we show that map recovery remains robust on MNIST image sequences, where each visit to a location yields a newly sampled image of its assigned digit. Across four heavily aliased environments, the end-to-end pipeline successfully uncovers the underlying adjacency graph with high edge precision and recall, directly from visual input. This work provides a proof of principle that CSCG can serve as a composable building block in a deep learning architecture.

AI 总结：研究连续时间生成框架中路径变形问题，提出有限时间谱灵敏度g(t)，通过跟踪状态转移矩阵奇异值揭示流几何。利用其在数据稀缺时的表现开发内部几何审计框架，能仅依内部轨迹动态检测生成记忆。

链接：https://arxiv.org/abs/2607.12616

作者：Shuchan Wang

英文摘要：Continuous-time generative frameworks construct probability paths between base and target domains by optimizing time-dependent velocity fields. While theoretical targets favor straight trajectories, empirical networks develop complex path deformations. This paper presents the Finite-Time Spectral Sensitivity (FTSS) g(t), a gradient-free, forward-pass metric that exposes flow geometry by tracking the root-mean-square singular value of the state-transition matrix. Serving as a continuous proxy for stable rank, g(t) reveals a distinct geometric pathology under data scarcity: while generalizing models maintain stable effective dimensions, overfitting causes a spectral collapse. We leverage this structural phenomenon to develop an internal geometric audit based on g(t). Our framework detects generative memorization using purely internal trajectory dynamics, removing the need for external membership queries or baseline data comparison.

AI 总结：研究视频扩散模型在多球动力学实验中的表现，发现其存在序列性差距，即任务所需串行计算与模型去噪循环不匹配，增加有效串行计算的方法可提升性能，还证明去噪步骤在串行推理和模拟任务上有结构障碍。

链接：https://arxiv.org/abs/2607.13031

机构：UC Berkeley（加州大学伯克利分校）

作者：Jorge Diaz Chao, Konpat Preechakul, Yuxi Liu, Yutong Bai

英文摘要：When one ball strikes another, then another, video models should predict the consequences of each bounce. In controlled experiments on multi-ball hard-sphere dynamics, we find that the performance of standard bidirectional video diffusion degrades as the causal chain lengthens, even when provided more denoising steps. In a length-matched single-ball control, where ball-ball interactions are absent, the degradation largely disappears, isolating dependent-event structure rather than video length as the cause. Across intervention studies, methods that increase effective serial computation improve performance disproportionately, including autoregressive/blockwise generation and architectural depth. We identify this pattern as the seriality gap: a mismatch between tasks requiring growing serial computation and video diffusion models whose denoising loop does not provide scalable serial compute. We then prove that, for deterministic video prediction, denoising steps do not add serial computation beyond the backbone, indicating a structural obstacle for video diffusion on serial reasoning and simulation tasks.

AI 总结：研究针对半监督学习方法依赖分布假设的问题，提出基于组件风险线性组合构建无偏风险估计器的广义框架，推导最小可实现方差，建立泛化界，并引入两种实用方法，在多分类基准测试中表现良好。

链接：https://arxiv.org/abs/2607.11947

机构：Institute of Science Tokyo（东京理科大学）； RIKEN Center for Advanced Intelligence Project（理化学研究所高级智能项目中心）

作者：Yushi Hirose, Hiroo Irobe, Takafumi Kanamori

英文摘要：Typical semi-supervised learning (SSL) methods rely on distributional assumptions, and their performance degrades when these are violated. While PNU learning, a risk rewriting method, offers a distribution-free alternative, it is restricted to binary classification and its variance optimality remains unclear. In this paper, we propose a generalized framework that constructs unbiased risk estimators using linear combinations of component risks, subsuming PNU learning and extending to multiclass classification. We derive the minimum achievable variance, demonstrating our estimator can attain lower variance than PNU in asymmetric loss scenarios. Furthermore, we establish a generalization bound directly linking this variance reduction to improved learning performance. Based on these theoretical insights, we introduce two practical SSL methods that empirically match or outperform existing approaches on binary and multiclass benchmarks.

AI 总结：研究网络对齐问题，提出FastAlign框架，它通过保留最优传输公式，结合稀疏感知图计算与特定领域内核融合，在不降低对齐精度的情况下，大幅提升了计算效率，减少了端到端运行时间。

链接：https://arxiv.org/abs/2607.11952

机构：Texas A&M University（德克萨斯农工大学）； Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）； University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）

作者：Elaheh Hassani, Durga Mandarapu, Qi Yu, Hanghang Tong, Ariful Azad

英文摘要： Network alignment identifies node correspondences across different networks and is a fundamental primitive in many data science applications, including social network analysis, fraud detection, and knowledge graph integration. However, state-of-the-art network alignment methods often achieve high accuracy by repeatedly constructing and updating dense matrices, sacrificing scalability in the process. To address this scalability limitation without compromising alignment accuracy, we present FastAlign, a scalable, sparsity-aware framework for optimal transport-based network alignment. Rather than introducing a new alignment model, FastAlign preserves the original OT formulation and reinterprets its computation as a set of recurring mixed sparse-dense operations. FastAlign combines sparsity-aware graph computation with domain-specific kernel fusion, including a custom SpMM kernel. Our results show that FastAlign achieves alignment quality comparable to state-of-the-art OT-based methods while substantially reducing end-to-end runtime up to 3.89x-9.45x on CPU and 2.24x-32.54x on GPU.

AI 总结：研究探讨用大语言模型进行零样本慢性肾病筛查的可行性，提出特征引导零样本框架，通过特征选择、序列化患者记录实现。实验表明该框架能提升模型性能，在多数据集有良好泛化性，为CKD筛查提供新方法。

链接：https://arxiv.org/abs/2607.12260

机构：School of Computing, Mathematics and Engineering, Charles Sturt University（查尔斯·斯特尔特大学计算、数学与工程学院）； Department of Computer Science, Rensselaer Polytechnic Institute（伦斯勒理工学院计算机科学系）

作者：Muhammad Ashad Kabir, Sirajam Munira

英文摘要：Early screening of chronic kidney disease (CKD) is essential for preventing irreversible progression; however, many machine learning (ML)-based screening methods remain difficult to deploy in community and resource-limited screening settings due to their reliance on large labeled datasets, resource-intensive pathology tests, or high-dimensional clinical features, and limited robustness to population and distributional shifts. This study examines the feasibility of using large language models (LLMs) for early-stage CKD screening in a zero-shot setting, without dataset-specific training. We propose a feature-guided zero-shot framework that evaluates LLM performance using a selected set of clinically meaningful, readily available community-based features, rather than exhaustive clinical inputs. Feature selection was guided by ML-based analysis to identify a compact, clinically relevant subset of variables. Tabular patient records were subsequently serialized into text using standardized prompt templates to enable zero-shot inference. The zero-shot performance of four LLMs (LLaMA-3, Qwen-3, Mistral, and GPT-4o-mini) was evaluated using both the full feature set and the selected subset. Generalizability was assessed across three heterogeneous CKD datasets spanning three countries. Across models and datasets, the selected feature set yielded consistent and statistically significant improvements in balanced accuracy and probability estimates, achieving performance levels suitable for screening purposes. These findings suggest that LLMs can support clinically meaningful, training-free CKD screening using minimal community-accessible patient features, offering a practical complement to conventional ML methods in real-world screening contexts.

AI 总结：研究热身 - 稳定 - 衰减学习率调度中冷却阶段效果，发现其取决于梯度噪声结构和优化器更新归一化。通过理论分析和模拟，揭示不同情况下冷却的作用机制，还在真实分类任务中验证噪声 regime 诊断，解释冷却是否有帮助。

链接：https://arxiv.org/abs/2607.12360

作者：Subham Singh, Ashutosh Mishra, Subha Raut

英文摘要：The cooldown phase of a warmup-stable-decay (WSD) learning-rate schedule, now a default in large-model pretraining, lowers the final training loss in some settings and does nothing in others. We give a provable account of which case obtains, and it turns on two properties together: the structure of the gradient noise and whether the optimizer normalizes its update. On a strongly convex objective with multiplicative (gradient-proportional) noise, stochastic gradient descent contracts geometrically at a constant learning rate, so cooldown has nothing to improve. Under the same objective and noise, sign-based and normalized methods, the standard surrogates for adaptive optimizers, settle on a noise floor of order $\eta^2$ and reach the minimizer only as the learning rate is driven to zero; any additive noise then reinstates a floor for every method. The mechanism is elementary: an SGD step shrinks in proportion to the gradient and so anneals itself, whereas a normalized step keeps unit scale and cannot. We solve the signSGD stationary law on the quadratic exactly and obtain the floor constant in closed form, prove a local form of the dissociation under $(L_0,L_1)$-smoothness, extend the floor to normalized SGD in dimension d>1 by a scale-invariance argument, and establish robustness to momentum and heavy-tailed noise. Simulation confirms every prediction, and we demonstrate the resulting noise-regime diagnostic on a real classification task with directly measured gradient noise. The mechanism explains whether cooldown helps; the interior cooldown fraction used at scale lies outside stationary landscape-and-noise geometry.

AI 总结：研究针对分子设计中搜索化学空间成本高的问题，提出样本高效生成优化（SEGO）框架，通过概率代理模型、生成模型等进行贝叶斯优化，在相关基准测试中大幅提升样本效率，接近由实验反馈驱动的优化。

链接：https://arxiv.org/abs/2607.12488

作者：Sarina Kopf, Cristina Nevado, Philippe Schwaller

英文摘要：Molecular optimization in drug discovery, materials design, and catalysis requires searching vast chemical spaces under tight evaluation budgets, since high-fidelity oracles and experimental measurements are costly. The practical impact of an optimization method therefore hinges on its sample efficiency: how few evaluations it needs to find strong candidates. We introduce Sample Efficient Generative Optimization (SEGO), a framework for Bayesian optimization on adaptively generated molecules. In SEGO, a probabilistic surrogate model forms a hypothesis about where hits lie in chemical space, a generative model is steered to propose candidates in that region, the most promising candidate is selected via an acquisition function, and the resulting oracle call is used both to sharpen the surrogate and to anchor the generator in real reward. SEGO attains state-of-the-art performance on the practical molecular optimization (PMO) benchmark using only one tenth of the oracle calls consumed by other methods, and on a multiparameter docking task it reaches ten hits in roughly half the oracle calls of existing approaches. These gains move molecular optimization closer to campaigns driven by direct experimental feedback.

AI 总结：研究刻画使表征先验起作用的因素，包括特征族、无标签不变性、时间窗口等方面。通过多种实验设置，发现特征族对齐影响泛化，无标签不变性可加速，早期短暂应用先验效果最佳，为理解和改进先验方法提供依据。

链接：https://arxiv.org/abs/2607.12735

作者：Gunner Levi Howe

英文摘要：Companion work showed the grokking delay is causally the time to form task-structured representations, injectable via a contrastive prior. Here we characterize what makes such a prior work, across four axes, in 188 new runs. Content: a coherent, learnable prior built from the wrong feature family (magnitude bands) blocks generalization like a random partition (1/15 vs 0/20 grok; $p=0.43$ between them), confirming the companion's prediction that priors act at the level of the circuit's features. Supervision: a fully label-free invariance prior -- positives are commuted pairs $(a,b)\sim(b,a)$ only -- generalizes in 15/15 runs at a median $2.7\times$ speedup, more reliably than the label-supervised prior itself ($p=0.038$), and combined with a weight-norm clamp yields the strongest method we test (median $17\times$, 5/5) -- strongest meaning reliably fast: plain cross-entropy with a clamp matches this speed only at the exact critical norm, while the prior keeps it fast across the entire clamp range. Timing: the prior is only needed early -- applied solely during the first 2000 epochs (4% of budget) it generalizes 10/10 at $2.7\times$, beating continuous application (8/10, $1.25\times$) and a duration-matched later window ($2.1\times$). Setting: the dissociation replicates on modular multiplication and across depths and normalization variants, and a clamp sweep quantifies the companion's central claim: structure injection flattens the weight-norm delay-law exponent about 17-fold (plain cross-entropy slows $31\times$ per +10 norm units, a lower bound as higher cells are censored, versus $1.22\times$ with the prior). Honest boundary: tasks that generalize before memorizing have no delay to control. Feature-family alignment decides whether a prior permits generalization; invariance content suffices for acceleration without labels; a brief early window captures nearly all of the benefit.

AI 总结：研究查询可见性对键值缓存压缩排名的影响，通过匹配预算审计六种压缩方法与三个基线，发现查询可见性改变排名，且方法在两种协议下的下降与问题对评分信号的可见度有关。

链接：https://arxiv.org/abs/2607.11942

机构：University of Technology Sydney（悉尼科技大学）

作者：Daming Luo, Christy Liang, Junyu Xuan

英文摘要：KV-cache compression methods are predominantly evaluated with the query appended to the context before compression -- a query-aware protocol. Yet the economic case for a compressed KV cache is reuse: compress a document once, answer many future questions against it. In that deployment, compression must happen query-agnostic -- before any question is seen. We present a matched-budget audit of six published compression methods against three trivial baselines on three open 7-9B models (144,300 paired evaluations on RULER-8192; 40,800 on LongBench; 50,000-resample paired bootstrap throughout). Everything is held fixed -- model, compression ratio, instances, decoding -- except the scoring rule. Three findings. (1) Query visibility changes the rankings: under the agnostic protocol, of the five audited methods that share a common attention backend, only KeyDiff beats a best-of-3 trivial baseline consistently (31 of 36 cells), and the most widely deployed method, SnapKV, loses to "keep the start and the recent window" on average (-0.066). (2) The per-method drop between the two protocols is ordered consistently with how visible the question is to each method's scoring signal, legible in its source code: from Delta=+0.198 for SnapKV (the question sits inside its 64-token observation window) down to Delta=+0.011 for KeyDiff (its score contains no query term at all).

AI 总结：研究针对KV缓存内存成本高的问题，提出JoLT方法，通过联合塔克分解和JL残差分配压缩缓存，单个拉格朗日对偶分配参数，实现近无损2 - 3倍压缩，随机变体FlashJoLT还加速了压缩时间。

链接：https://arxiv.org/abs/2607.12550

机构：Universität Trier（特里尔大学）； Fachbereich IV, Mathematik, Universität Trier（特里尔大学数学四系）

作者：Rahul Krishnan, Volker Schulz

英文摘要：The key-value (KV) cache has become the dominant memory cost of transformer inference. It grows with batch size, context length, and depth, and at long context it, rather than the model weights, sets the ceiling on throughput. Two families of methods reduce it. Low-rank methods factor two-dimensional slices of the cache, either per-head matrices or cross-layer feature blocks, and quantization methods lower the bit-width of every entry. Neither family exploits the fact that the cache at a layer is naturally a third-order tensor whose three axes, the heads, the tokens, and the features, carry very different amounts of redundancy. We take this tensor view directly. Our method, JoLT, applies a partial Tucker decomposition that compresses only the token and feature axes while leaving the head and layer axes intact, and then restores the energy that truncation discards with a Johnson-Lindenstrauss (JL) rotated low-bit residual. A single Lagrangian dual allocates the Tucker ranks and the residual bit-widths together, per layer group and separately for keys and values, under one byte budget. The result is a near-lossless 2-3x compression: perplexity, GSM8K accuracy, and RULER needle-in-a-haystack retrieval all stay at or within statistical noise of the uncompressed baseline on both a grouped-query-attention model (Mistral-7B-v0.3) and a multi-head-attention model (LLaMA-2-13B). At 2x, JoLT reconstructs the cache to relative Frobenius error 0.009 (K) and 0.006 (V) on both architectures, roughly an order of magnitude below cross-layer SVD and 4-bit quantization. A randomized-SVD variant, FlashJoLT, delivers a 5-13x compression-time speedup at matched quality.

AI 总结：综述介绍用于扩散大语言模型的统一延迟分解框架，以理清算法、架构和系统因素对推理速度的影响，将加速技术分类，提供可重复基准测试指导方针并强调实现并行生成潜力的挑战。

链接：https://arxiv.org/abs/2607.12829

机构：KAIST AI（韩国科学技术院人工智能研究所）； Yonsei University（延世大学）

作者：Daehoon Gwak, Minhyung Lee, Junwoo Park, Jaegul Choo

英文摘要： Diffusion large language models (dLLMs) offer a theoretical advantage in parallel generation over standard autoregressive models. However, parallel generation alone does not guarantee practical speedups. Realizing this efficiency requires specialized inference mechanisms, such as diffusion-aware caching and reuse. Consequently, as inference efficiency becomes a prerequisite for practical deployment, recent research has actively explored acceleration techniques across algorithms, architectures, and systems. However, rigorous comparisons remain difficult, as end-to-end latency stems from intricate trade-offs between algorithmic, architectural, and system-level factors that are often conflated in existing benchmarks. In this survey, we introduce a unified latency decomposition framework for dLLMs to disentangle these factors and analyze their impact on inference speed in real deployments. Guided by this framework, we categorize acceleration techniques along three axes covering algorithmic innovations, architectural and system optimizations, and inference-time scaling. Finally, we provide guidelines for reproducible benchmarking and highlight open challenges for realizing the full potential of parallel generation.

AI 总结：针对当前模型遗忘方法缺乏精确引导信号的问题，提出GSUO框架，设计特定任务细粒度引导信号指导遗忘过程，适用于多种遗忘任务。实验表明，该框架在遗忘有效性、泛化、效率等方面表现出色，优于14个基线。

链接：https://arxiv.org/abs/2607.11975

作者：Xujia Li, Dan Li, Jian Lou, Wenjie Feng

英文摘要：Current machine unlearning methods predominantly rely on global, coarse-grained intervention strategies. They lack precise pilot signals to guide the unlearning process and fail to provide differentiable guidance across different unlearning tasks. Due to the varying memorization strengths of samples during original training, such a uniform strategy leads to two problems: some samples are over-unlearned, which harms model utility; while others are under-unlearned, leaving residual information that can be exploited by privacy attacks. In this paper, we propose GSUO, a guidance-signal-aware unlearning optimization framework that designs task-specific fine-grained guidance signals to steer the unlearning process and is applicable to both random-subset and class-wise forgetting tasks. Extensive experiments demonstrate that GSUO outperforms 14 baselines in terms of both unlearning effectiveness and generalization, while achieving high efficiency and significant speedups, validating its effectiveness for reliable machine unlearning.

AI 总结：研究远距离单次条纹投影轮廓术存在的形状先验捷径问题，提出PhiCalNet方法，通过输出包裹相位表示并映射到深度，消除形状先验解决方案，降低物体平均绝对误差，提升了轮廓术性能。

链接：https://arxiv.org/abs/2607.11928

机构：Department of Mechanical Engineering, Iowa State University（爱荷华州立大学机械工程系）； College of Engineering, University of Georgia（佐治亚大学工程学院）

作者：Adam Haroon, Cody Fleming, Beiwen Li

英文摘要：Single-shot fringe projection profilometry (FPP) networks that regress depth directly can exploit a shape-prior shortcut, recovering depth from object boundaries rather than from fringe phase. On a photorealistic synthetic benchmark (15,600 fringe images, 50 objects at 1.5-2.1 m standoff), the best such UNet baseline plateaus at 14.54 mm object mean absolute error (MAE), and neither more data nor more capacity removes the shortcut, because neither changes the hypothesis space the optimizer searches. We introduce PhiCalNet, which outputs a wrapped-phase representation $(\sin\phi, \cos\phi)$ and maps it to depth through a fixed differentiable calibration layer, removing the shape-prior solution architecturally rather than by a loss penalty. Because the single-shot mapping is non-injective without fringe order, PhiCalNet takes the fringe order as auxiliary input, an assumption a sensitivity analysis shows tolerates realistic decoding error; a physics-informed (PINN) baseline with the same physics as a soft penalty yields no gain, isolating the architectural choice as the operative factor. PhiCalNet reduces object MAE 3.3x to 4.46 mm, its residual confined to 0.103% of pixels at the $\pm\pi$ wrap discontinuity, and a three-frame extension reaches 1.16 mm. Two checks agree: interpretability makes phase the most decodable internal feature, and pixel-wise conformal uncertainty quantification, to our knowledge the first for FPP, localizes error at the same discontinuity, where rejecting the top 5% of pixels by snapshot disagreement cuts root-mean-square error by 64% versus 3.5% for the baseline.

AI 总结：研究针对长周期电池退化模拟计算昂贵的问题，提出BattVAE - GP框架，先利用VAE编码数据至潜在空间，再用GP训练并估计不确定性，能准确恢复轨迹，提供高效且具不确定性意识的长周期退化建模代理。

链接：https://arxiv.org/abs/2607.11943

机构：CNRS（法国国家科学研究中心）

作者：Raghvender Raghvender, Mahdi Abid, Ferran Brosa Planella, Charles Delacourt, Arnaud Demortière

英文摘要： Long-horizon physics-based simulations of battery degradation provide mechanistic insight but remain computationally expensive, limiting their use for dense exploration of operating conditions over extended cycle life. Here, we propose a hybrid physics-probabilistic learning framework for surrogate modeling of lithium-ion battery degradation trajectories at unseen charging rates. Cycle-resolved degradation data generated with a DFN/P2D electrochemical model in PyBaMM are first transformed into capacity-aligned voltage and derivative features and encoded using a Variational Autoencoder (VAE). The resulting two-dimensional latent space organizes degradation trajectories according to both cycle progression and charging protocol. A sparse multitask Gaussian process (GP) is then trained in this latent space using cycle number and C-rate as input variables, providing continuous interpolation of latent degradation dynamics together with posterior uncertainty estimates. Under protocol-level holdout evaluation, the latent-space GP accurately recovers unseen C-rate trajectories and exhibits uncertainty behavior consistent with the support of the training data. When queried at unseen interior C-rates, the model generates latent trajectories that remain coherently positioned between neighboring simulated protocols. Decoding the GP-predicted latent states through the frozen VAE decoder yields smooth voltage-capacity evolution, while Monte Carlo propagation of the GP latent posterior through an auxiliary latent to State of Health (SOH) predictor provides uncertainty-aware SOH estimates. The proposed BattVAE-GP framework therefore offers a computationally efficient and uncertainty-aware surrogate for long-horizon degradation modeling, providing a structured basis for extending battery health prediction toward richer operating conditions and future simulation-experiment fusion.

AI 总结：该研究对慢性肾病预测的机器学习模型进行系统综述，引入结构化分类法和评分框架评估可靠性，发现泄露与性能夸大有关，多数预测器缺乏稳定性，揭示很多性能提升源于方法局限而非真实预测力。

链接：https://arxiv.org/abs/2607.11963

机构：East West University（东西方大学）

作者：Mashrul Hossain, Nafesa Kibria, Fahim Shahriar

英文摘要：The early detection of Chronic Kidney Disease using machine learning has attracted significant interest in healthcare-related computer science. Despite rapid advancements in this field, many reported studies remain inconsistent and potentially misleading. A significant drawback is the lack of organized evaluation regarding methodological concerns. Key issues include data leakage, limited access to temporal patient records and inconsistency in reported clinical indicators. This research offers a systematic literature review of existing CKD prediction studies using interpretable machine learning techniques, where nineteen relevant studies were selected via systematic searches across major academic databases. To assess methodological reliability, this study introduces a structured taxonomy of information leakage and a quantitative leakage scoring framework to systematically evaluate reliability across CKD prediction studies. The analysis reveals a strong relationship between leakage and inflated performance. Here, High leakage-studies report an average accuracy of 95.48%, compared to 80.2% for leakage-free studies, reflecting an increase of approximately 15.28%. Furthermore, a cross-study feature stability analysis shows that only a small subset of predictors is consistently reproducible, with over 80% lacking reliability. Overall, the findings suggest that many reported performance improvements stem from methodological limitations rather than true predictive capability.

AI 总结：研究如何可靠检测分布外样本，利用稀疏自编码器从神经网络中间激活中学习可解释特征，通过余弦相似度得出OOD分数，该方法在检测基准上性能领先且能提供可解释见解。

链接：https://arxiv.org/abs/2607.12094

机构：Purdue University（普渡大学）； EPFL（洛桑联邦理工学院）

作者：Ayush Karmacharya (1), Luke Luschwitz (1), Lucia Romero (1), Yanan Niu (2), Joseph Campbell (1) ((1) Purdue University, (2) EPFL)

英文摘要：Reliable detection of out-of-distribution (OOD) samples is crucial for the safe deployment of machine learning models. Neural networks often produce overconfident predictions for inputs that deviate from their training data, leading to significant degradation in performance. While many OOD detection methods focus on the final output layer, they neglect the rich hierarchical information present in intermediate network layers. This paper introduces a novel approach that leverages sparse autoencoders (SAEs) to learn interpretable features from these intermediate activations. We find that in-distribution (ID) and OOD data activate distinct sets of these sparse features. We propose a new OOD score derived from the cosine similarity between the sparse feature activations of a test sample and the mean activations of ID classes. Our post-hoc detection method not only achieves state-of-the-art performance on standard OOD detection benchmarks, but yields interpretable insights into how distribution shift affects learned representations.

AI 总结：挑战信息依赖导致训练数据隐私泄露的观点，通过三个结果揭示对抗性非鲁棒特征才是原因，引入反对抗训练建立因果关系并揭示隐私-鲁棒性权衡，修正对训练数据暴露的理解。

链接：https://arxiv.org/abs/2607.12354

机构：Dartmouth College（达特茅斯学院）； Breuer Lab（布鲁尔实验室）

作者：Rasmus Torp, Shailen K. Smith, Adam Breuer

英文摘要：In this paper, we challenge the prevailing view that information dependency (including rote memorization) drives training data exposure to image reconstruction attacks. We show that extensive exposure can persist without rote memorization and is instead caused by a tunable connection to adversarial robustness. We begin by presenting three surprising results: (1) recent defenses that inhibit reconstruction by Model Inversion Attacks (MIAs), which evaluate leakage under an idealized attacker, do not reduce standard measures of information dependency (HSIC); (2) models that maximally memorize their training datasets remain robust to MIA reconstruction; and (3) models trained without seeing 97% of the training pixels, where recent information-theoretic bounds give arbitrarily strong privacy guarantees under standard assumptions, can still be devastatingly reconstructed by MIA. To explain these findings, we provide causal evidence that privacy under MIA arises from what the adversarial examples literature calls ``non-robust'' features (generalizable but imperceptible and unstable features). We further show that recent MIA defenses obtain their privacy improvements by unintentionally shifting models toward such features. To establish this causal relationship, we introduce Anti Adversarial Training (AT-AT), a training regime that intentionally learns non-robust features to obtain both superior reconstruction defense and higher accuracy than state-of-the-art defenses. Our results revise the prevailing understanding of training data exposure and reveal a new privacy-robustness tradeoff.

AI 总结：研究在线手写识别模型的对抗攻击问题，提出基于显著性引导时间编辑的框架，通过按时间显著性插入和删除点生成对抗样本，在保持手写视觉结构的同时具有更强的一次性黑盒转移性。

链接：https://arxiv.org/abs/2607.12500

机构：Kyushu University（九州大学）

作者：Yataro Tamura, Brian Kenji Iwana, Jiseok Lee

英文摘要： Deep learning models for online handwriting recognition have been shown effective and are increasingly deployed in practical applications. However, their vulnerability to adversarial attacks is still a challenge. Existing adversarial methods are predominantly designed for image-based inputs and typically rely on additive spatial perturbations. When applied to online handwriting, which is inherently represented as a time series of pen trajectories, such perturbations often introduce high-frequency jitter and visibly unnatural stroke artifacts. In this work, we propose a novel adversarial attack framework for online handwriting recognition based on salience-guided temporal editing. Instead of adding noise, the proposed method generates adversarial examples by inserting and deleting points at time steps selected according to temporal salience, preserving the shape and smoothness of the original handwriting. Temporal salience is estimated using gradient-based activation mapping, which guides edits toward time steps that strongly support the original class prediction. We evaluate the proposed approach on the Unipen and CASIA-OLHWDB datasets under both white-box and one-shot black-box attack settings. Experimental results demonstrate that while conventional image-based attacks achieve strong white-box performance, they exhibit poor transferability across models. In contrast, the proposed temporal editing attack achieves stronger one-shot black-box transferability while preserving the visual structure of the handwriting. These results indicate that temporal editing is a relevant threat model for online handwriting recognition, particularly in one-shot black-box transfer settings.

AI 总结：该研究针对强化学习中分布外检测问题，提出OOD-RL-Bench框架，通过共享接口集成检测器与异常注入器，在LunarLander-v3环境评估其效用，揭示不同异常类型性能差异，还公开相关成果以便重现。

链接：https://arxiv.org/abs/2607.12523

机构：Deakin University（迪肯大学）； Federation University Australia（澳大利亚联邦大学）

作者：Emil Mittag, Richard Dazeley, Peter Vamplew

英文摘要：Reliable reinforcement learning (RL) agents must maintain operational integrity amidst sensor malfunctions, dynamic disturbances, and slow environmental shifts. The detection of out-of-distribution conditions is pivotal to determining when an agent's observations, transitions, or trajectory dynamics deviate from the assumptions underpinning its policy training. Current out-of-distribution (OOD) detection benchmarks typically evaluate image classifiers or static low-dimensional datasets, failing to account for the complex, action-dependent temporal structure inherent in RL trajectories. To address this gap, we present OOD-RL-Bench, a comprehensive and extensible framework designed to evaluate OOD detectors against categories of anomalies injected into RL trajectories. Detectors and anomaly injectors are integrated through shared interfaces and configuration, which allows new scoring methods and perturbation families to be evaluated without modification of the core benchmark loop. We evaluate the utility of the framework using a Deep Q-Network policy within the LunarLander-v3 environment. We assess the performance of each detector across a suite of anomaly types using matched-time AUROC, matched-time AUPRC, matched-time false-positive rate, detection delay, and segmented-onset metrics. Our analysis reveals significant performance variance across anomaly types: observation perturbations and regime switches are identified with high accuracy by several methods, while observation delay and action-conditioned dynamics remain difficult even when post-onset anomaly scores are compared against clean scores from the same timesteps. We make the framework, trained policy checkpoint, and complete results publicly available as a reproducible artefact.

AI 总结：研究在线二元序贯校准问题，基于已有突破成果，提出结合\textsc{SPR - 校准}程序与外部校正层的随机预测器，实现\(O(T^{2/3 - ε})\)期望校准误差，并对总校准误差进行分解及控制。

链接：https://arxiv.org/abs/2607.12928

机构：HKUST（香港科技大学）

作者：Zihan Zhang

英文摘要：We study the online binary sequential calibration problem. A recent breakthrough by \citet{dagan2024breaking} overcomes the classical \(T^{2/3}\) barrier for calibration error. Building on this result, we present an efficient randomized forecaster that achieves an expected calibration error \(O(T^{2/3-\varepsilon})\) for some constant \(\varepsilon>0\). Our forecaster combines the \textsc{SPR-Calibration} procedure \citep{dagan2024breaking} with an outer Blackwell-style correction layer. The \textsc{SPR-Calibration} procedure controls calibration with respect to a surrogate sequence of conditional-mean estimates, while the correction layer controls the additional error incurred when these surrogates are used to approximate the true outcomes. The analysis decomposes the total calibration error into the surrogate calibration error and the residual discrepancy between the surrogate sequence and the true outcomes. The former is bounded by the \textsc{SPR-Calibration} guarantee in \citet{dagan2024breaking}, and the latter is controlled using a quadratic potential argument together with the sparsity of the \textsc{SPR-Calibration} forecaster.

AI 总结：研究旨在弥合离散和网格化的PM10预测。核心方法是基于卷积条件神经过程的OmniPM-Net，融合化学传输模型与图神经网络预测。主要贡献是提升预测准确性，降低误差，在高浓度尾部和沙尘事件中有显著改进，还能提供网格化场。

链接：https://arxiv.org/abs/2607.11896

机构：ColorfulClouds Technology Co., Ltd.（彩云科技有限公司）； School of Systems Science, Beijing Normal University（北京师范大学系统科学学院）； D-ITET, ETH Zurich（苏黎世联邦理工学院动态信息与电气工程系）

作者：Shuangshuang He, Shuo Wang

英文摘要： Forecasting particulate matter (PM10) requires both station-scale accuracy and continuous spatial fields, especially during severe dust storms. Chemical transport models (CTMs) provide gridded forecasts but retain local biases, whereas graph neural networks (GNNs) track monitoring sites well at short lead times but do not produce gridded outputs. Here we present OmniPM-Net, a Convolutional Conditional Neural Process (ConvCNP)-based fusion model that reconciles these two forecast types within a shared spatial representation. A terrain-aware Gaussian set convolution lifts irregular GNN station forecasts onto a regular grid, where a multi-scale Spatial Source Attention (SSA) module blends them with Copernicus Atmosphere Monitoring Service (CAMS) forecasts; a shared omni-query readout then decodes this representation into consistent PM10 predictions at either stations or grid cells over a 108 h horizon. Evaluated across 1,618 air-quality monitoring stations throughout China over the full year of 2024, OmniPM-Net matches the station-level accuracy of the stronger GNN baseline (mean absolute error 21.14 versus 22.00 ug/m3) and reduces the CAMS mean absolute error by 30%, while simultaneously delivering the gridded fields that the discrete GNN cannot. Its clearest gains are in the high-concentration tail, where the 90th-percentile MAE falls by 9% relative to the GNN and 25% relative to CAMS, and during dust episodes, where it improves categorical detection skill while tracking the evolving spatial trajectory.

AI 总结：研究利用时间图机器学习为机构股票持有预测任务引入基准，将其作为节点亲和力预测，在含99名经理和标准普尔500指数成分股的数据集上，NAVIS模型表现出色，优于其他动态图和启发式方法，为时间图机器学习在相关领域提供可重复基础。

链接：https://arxiv.org/abs/2607.12067

机构：Amirkabir University of Technology（伊朗阿米尔卡比尔理工大学）

作者：Emad Izadifar, Zahed Rahmati

英文摘要：Institutional equity holdings disclosed in SEC Form 13F filings provide a rich temporal record of portfolio decisions by large investment managers. However, forecasting future allocations and modeling future demand remains challenging due to disclosure lags, reporting noise, and strong persistence in institutional behavior. We introduce the first benchmark for these tasks using temporal graph machine learning, framing holdings prediction as node affinity prediction -- i.e., forecasting portfolio weights -- on a discrete-time temporal bipartite graph of managers and securities extracted from preprocessed filings. On a sampled dataset comprising 99 managers and the S\&P 500 index (503 securities, 209,351 temporal edges across 48 quarters from 2013--2025), Node Affinity prediction model using Virtual State (NAVIS) achieves a state-of-the-art test Normalized Discounted Cumulative Gain (NDCG) of 0.9127 with features (0.9121 without), outperforming all dynamic graph representation learning competitors by a substantial margin, and outperforming all heuristic methods. Remarkably, a simple Exponential Moving Average baseline achieves 0.8882, surpassing all dynamic graph models except NAVIS and all heuristics except Persistent Forecast (0.8891), highlighting the strong smoothness and persistence of institutional portfolios. Domain-specific node features provide only marginal gains (<1.2\%), indicating that temporal and structural signals in the 13F ownership graph already capture most of the predictable information. By benchmarking a suite of Temporal Graph Benchmark (TGB) models under the node affinity prediction setting, both with and without features, on real-world 13F data, this work provides a reproducible foundation for temporal graph machine learning in holdings prediction and portfolio allocation.

AI 总结：介绍量子端口哈密顿神经网络，其基于同构哈密顿映射，通过测量诱导非线性学习经典动力学，在四种架构中实例化，经非线性摆和阻尼谐振子实验，展示了能量漂移、单调性及阻尼系数识别等方面的成果。

链接：https://arxiv.org/abs/2607.12269

机构：Mindverse Computing LLC（Mindverse计算有限责任公司）

作者：Dibakar Sigdel

英文摘要：We introduce Quantum Port-Hamiltonian Neural Networks (Q-pHNNs), a family of parameterised quantum circuits that learn classical dynamics in a structure-preserving manner. The framework relies on the Isomorphic Hamiltonian Mapping (IHM): the skew-symmetric interconnection matrix $\mathbf{J}$ corresponds to unitary gate evolution, and the positive-semidefinite dissipation matrix $\mathbf{R}$ corresponds to Measurement-Induced NonLinearity (MINL) realised via mid-circuit measurement and classical feedforward. This ensures conservation and passivity are enforced by construction rather than penalty terms. We instantiate the IHM in four architectures: (1) a Quantum HNN that learns conservative energy manifolds and extracts Hamilton's equations exactly via the Parameter-Shift Rule; (2) a Q-pHNN using Born-rule measurement for dissipation; (3) a Q-pHNN jointly learning the energy ansatz and damping coefficient; and (4) a topology-entangled Quantum Graph Neural Network for $N$-node coupled-phasor networks. Experiments on the nonlinear pendulum and damped harmonic oscillator demonstrate: (i)~$1.35\%$ relative energy drift with a symplectic integrator and scale correction; (ii)~$100\%$ energy monotonicity for the MINL circuit; and (iii)~$12.1\%$ error in damping-coefficient identification from vector-field snapshots with no direct supervision on the damping coefficient.

AI 总结：研究在有限内存下微调大型预训练模型的问题，提出CARE-LoRA框架，利用LoRA投影结构，用低秩压缩激活取代完整输入激活，并计算重建矩阵，大幅减少内存占用，性能与标准LoRA及变体相当甚至更优。

链接：https://arxiv.org/abs/2607.11940

机构：Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University（上海交通大学图像处理与模式识别研究所）

作者：Gengyu Zhang, Haiyin Ran, Zhengbao He, Yuhang Liu, Hanling Tian, Zhehao Huang, Xiaolin Huang

英文摘要：As the scale of large pre-trained models continues to grow, fine-tuning them under limited memory budgets has become increasingly challenging. Low-Rank Adaptation (LoRA), currently one of the most widely adopted parameter-efficient fine-tuning (PEFT) methods, mitigates this challenge by optimizing only low-rank adaptation matrices, thereby greatly reducing the number of trainable parameters. With the parameter overhead substantially reduced, the activations retained for backpropagation have emerged as the primary remaining memory bottleneck during LoRA fine-tuning. To address this, we propose CARE-LoRA, a data-aware Compressed Activation REconstruction framework. By exploiting the inherent projection structure of LoRA, CARE-LoRA replaces the full input activation with the low-rank compressed activation naturally produced by the LoRA branch. It further computes a lightweight reconstruction matrix during the forward pass with negligible additional computation cost, which is used during backpropagation to reconstruct the gradient signal, thereby keeping LoRA matrices fully trainable. Extensive experiments across diverse models and downstream tasks demonstrate that, while substantially reducing the overall memory footprint, CARE-LoRA achieves competitive or even superior performance compared with standard LoRA and representative LoRA variants. Our code is publicly available at this https URL.

AI 总结：针对分布式网络中联邦微调平衡全局知识与局部适应的挑战，提出PFAdapter框架，用分层LoRA分解将适配器参数分离，通过正交正则化和选择性聚合协议减少通信成本，实验证明该框架在多数据集上性能优于基线，为智能AI部署提供有效方案。

链接：https://arxiv.org/abs/2607.12111

机构：College of Future Information Technology, Fudan University（复旦大学未来信息技术学院）； Division of Natural and Applied Sciences, Duke Kunshan University（昆山杜克大学自然科学与应用科学部）； Department of Electrical and Computer Engineering, The University of British Columbia（英属哥伦比亚大学电气与计算机工程系）； Ant Group（蚂蚁集团）； College of Information Science and Electronic Engineering, Zhejiang University（浙江大学信息科学与电子工程学院）； School of Data Science and Engineering, East China Normal University（华东师范大学数据科学与工程学院）； College of Intelligent Robotics and Advanced Manufacturing, Fudan University & Fysics AI（复旦大学智能机器人与先进制造学院及复肆智能科技（上海）有限公司）； Xiaomi EV, Xiaomi Campus（小米汽车、小米园区）； Information and Communications Technology Cluster, Singapore Institute of Technology (SIT)（新加坡理工学院信息通信技术集群）； College of Electronic and Information Engineering, Tongji University（同济大学电子与信息工程学院）； School of Computer Science and Informatics, Cardiff University（卡迪夫大学计算机科学与信息学院）

作者：Jing Liu, Kun Yang, Yan Wang, Dingkang Yang, Xiaoshuai Hao, Wei Zhang, Yang Liu, Wei Zhou

英文摘要：Agentic AI systems are reshaping communications and networking by deploying autonomous intelligent agents capable of collaborative learning while maintaining data privacy at network edges. Within distributed network environments, Multimodal Large Language Models (MLLMs) serve as cognitive engines for edge devices, yet federated fine-tuning faces substantial challenges in balancing global knowledge aggregation with local adaptation under heterogeneous network conditions. Conventional federated protocols typically rely on uniform parameter aggregation, which conflates domain-invariant features with client-specific nuances, thereby resulting in suboptimal personalization and excessive communication overhead. To address these challenges, we propose PFAdapter, a communication-efficient framework introducing hierarchical LoRA decomposition to explicitly separate adapter parameters into global-shared and local-private components. Query and key projections are assigned to global synchronization for capturing universal multimodal semantics across the network, while value and output projections remain localized for edge-specific adaptation. Additionally, orthogonality regularization based on the Frobenius norm enforces strict separation between these components, preventing redundant feature learning. Selective aggregation protocols synchronize only global-shared components across the federated network, preserving local expertise and reducing communication costs by nearly 50%. Extensive experiments on VQA-RAD, SLAKE, Hateful Memes, and CrisisMMD datasets demonstrate that PFAdapter consistently outperforms state-of-the-art baselines, achieving accuracy improvements ranging from 2.4% to 4.8% across diverse edge intelligence tasks. Consequently, our framework establishes an efficient solution for agentic AI deployment in resource-constrained communication networks.

AI 总结：研究针对联邦多模态大语言模型微调中灾难性遗忘问题，提出FedCMM框架，在参数、数据、聚合三个层面嵌入持续学习保障，经实验验证该框架在准确性和反向迁移上优于基线，能实现跨异构网络AI部署的稳健进化适应。

链接：https://arxiv.org/abs/2607.12112

机构：The University of British Columbia（英属哥伦比亚大学）； Fudan University（复旦大学）； Royal College of Science, Imperial College London（伦敦帝国理工学院皇家科学学院）； Dyson School of Design Engineering（戴森设计工程学院）； Suzhou Institute of Biomedical Engineering and Technology (SIBET), Chinese Academy of Sciences（中国科学院苏州生物医学工程技术研究所）； East China Normal University（华东师范大学）

作者：Jing Liu, Chenxuanyin Zou, Jiayang Ren, Gaoyun Fang, Chengfang Li, Yan Wang, Zhenchao Ma, Bo Hu

英文摘要：Federated fine-tuning of Multimodal Large Language Models (MLLMs) across distributed networks enables privacy-sensitive adaptation to evolving data streams, yet a fundamental obstacle prevents robust deployment in dynamic environments: catastrophic forgetting, wherein sequential task updates erase previously acquired knowledge across visual, linguistic, and cross-modal representations. Addressing this challenge is especially critical for autonomous networked AI operating in safety-sensitive domains, such as content moderation, where reliable retention of prior knowledge underpins system integrity. To overcome this, we propose Federated Continual Multimodal Learning (FedCMM), a framework that embeds continual-learning safeguards into the federated optimization loop at three complementary levels. At the parameter level, modality-aware elastic weight consolidation computes separate Fisher information matrices for the vision encoder, language backbone, and cross-modal projector, providing granular, asymmetry-aware protection against modality-specific forgetting. At the data level, each client trains a lightweight local generative replay module to synthesize raw-data-free embedding-level multimodal replay tuples without any raw data sharing. At the aggregation level, Task-similarity-aware gradient aggregation autonomously filters and reweights client updates by gradient cosine similarity, suppressing conflicting directions and stabilizing the global learning trajectory. Extensive experiments on two benchmarks demonstrate that FedCMM consistently outperforms recent baselines on accuracy and backward transfer, confirming that holistic, modality-aware optimization enables robust evolutive adaptation across heterogeneous networked AI deployments.

AI 总结：研究利用大语言模型学习化学反应机理推理，构建大规模推理数据集及福山基准，通过机理感知训练微调Qwen3 - 30B - A3B，在福山基准集A上精确路径匹配超FlowER模型，增强了语言模型的化学推理能力。

链接：https://arxiv.org/abs/2607.12771

机构：Princeton University（普林斯顿大学）； University of Pittsburgh（匹兹堡大学）； Northeastern University（东北大学）； University of Florida（佛罗里达大学）

作者：Xingyu Dang, Haocheng Tang, Junmei Wang, Yanjun Li

英文摘要： Reaction mechanisms consist of the step-by-step sequences of elementary reactions that explain chemical transformations. Learning the mechanism logic is therefore essential for enhancing the fundamental chemical intelligence of large language models (LLMs). The stepwise deduction of reaction mechanism aligns naturally with the reasoning paradigms of reasoning LLMs. However, current chemical LLMs primarily emphasize coarse-grained name reactions for product prediction and retrosynthesis, often leading to physical inconsistencies and hallucinations. In contrast, specialized small-scale generative models for mechanism inference typically suffer from restricted generalization capacity across diverse chemical spaces. To overcome these limitations, we built a novel, large-scale reasoning dataset of reaction mechanisms. Furthermore, we established the FukuyamaBench, a difficult benchmark derived from Fukuyama's Advanced Organic Reaction Mechanism book, to rigorously evaluate model performance on hierarchical mechanism reasoning. Our fine-tuned Qwen3-30B-A3B achieves 8.3% exact pathway match on FukuyamaBench Set~A, surpassing the specialized FlowER model (5.1%), demonstrating that mechanism-aware training substantially enhances chemical reasoning in language models.

AI 总结：研究针对现有 GPU 索引器 - TopK 内核效率低的问题，利用维度诅咒设计 LITETOPK 内核，通过采样估计得分范围划分候选结果桶，降低内存开销，在实际部署中加速 GLM 5.2 预填充阶段，提升了效率。

链接：https://arxiv.org/abs/2607.11976

机构：Nanyang Technological University（南洋理工大学）； ETH Zurich（苏黎世联邦理工学院）； The Chinese University of Hong Kong（香港中文大学）

作者：Ziqi Yin, Jianyang Gao, Peiqi Yin, Jiangneng Li, Gao Cong

英文摘要：Indexer-TopK, the operation to compute the scores and select the top-k candidates, is widely used by sparse attention kernels in large language models and vector retrieval in recommendation systems and vector databases. However, existing GPU-based Indexer-TopK kernels like DeepSeek Sparse Attention (DSA) remain inefficient due to excessive global memory traffic, costly synchronization, and prohibitive memory overhead. In this work, we exploit the curse of dimensionality in high-dimensional spaces, where distances between high-dimensional vectors tend to concentrate within a narrow range, to design LITETOPK, a novel and efficient fused Indexer-TopK kernel. LITETOPK first samples a small subset of data to estimate query-data score ranges, then uses these estimates to partition candidate results into bins online. This organization allows the LITETOPK kernel to maintain a tight approximate threshold, write back only promising candidates, reduce unnecessary I/O, substantially lower memory overhead, and still preserve exact Top-k correctness. Experimental results show that LITETOPK accelerates the prefill stage of GLM 5.2 by 1.2x in real-world deployment scenarios while incurring lower memory overhead.

AI 总结：该研究基于人工智能科学社区提出智能体方法用于自动神经算子发现，通过虚拟实验室中三个智能体协作，共享通用词汇表，在五个问题上评估，能发现高精度低参数架构，揭示语言模型智能体对保持多样性的作用及神经算子无免费午餐定理。

链接：https://arxiv.org/abs/2607.12122

机构：Texas A&M University（德克萨斯农工大学）

作者：Luis Loo, Ulisses Braga-Neto

英文摘要：We present an agentic approach to autonomous neural operator discovery based on an AI scientific community, which consists of a swarm of virtual laboratories that interact under a citation-based economy of influence. Highly-cited labs found new labs that follow their research direction and replace non-performing labs. Each virtual lab contains three agents: an LLM planner that proposes an architecture, a numerical worker that trains and measures it, and an LLM reviewer that participates in cross-lab peer review. All labs share a common vocabulary consisting of DeepONet (branch-trunk), Fourier, Transformer (attention), wavelet, and residual convolutional neural operator building blocks. We evaluate the neural operator AI scientific community on five problems, namely piecewise regression, the linear advection and Burgers 1D PDEs, and the Navier-Stokes and Darcy flow 2D PDEs, while repeating the simulation three times for each problem. The results show that the neural operator AI scientific community is capable of discovering high-accuracy, low-parameter-count neural operator architectures. All 9,623 LLM calls are logged and audited, which reveals that the virtual lab LLM planners choose to hybridize in 99.8% of their logged decisions, consistently returning multi-family hybrids. Moreover, we conducted an ablation study by replacing the LLM agents in each lab by rule-based alternatives, which caused the scientific community to collapse to non-hybridized single-family stacks in several cases, showing that LLM agency is needed to preserve diversity. The results suggest a no-free-lunch theorem for neural operators: there is no universal winner. The code, configurations, and the complete LLM transcripts are released at this https URL.

AI 总结：研究以钻孔换热器为奇异源的非均质土壤热问题，开发参数化物理信息神经网络，利用解析线源模型等特性，通过温度变化分解等步骤将原始问题重述，经训练得到通用校正器，数值测试验证了该方法的有效性。

链接：https://arxiv.org/abs/2607.12271

机构：Technical University of Munich（慕尼黑工业大学）

作者：Moke Rao, Thomas Hamacher, Smajil Halilovic

英文摘要：In this paper, a parametric physics-informed neural network for solving the heterogeneous soil thermal problem with borehole heat exchangers (BHEs) as singular sources is developed. There are three novel features in the present framework; namely, (i) the singularity is naturally removed by using analytical line source models; (ii) using the explicit formulation for gradient thermal conductivity enables physics-informed learning of the parametrization featuring the conductivity; (iii) the learned correction is utilized as an efficient universal corrector via superposition principles. We first introduce the decomposition of the temperature change and transform the approximation of the entire heterogeneous response to the correction compensating the difference between the practical solution and idealized homogeneous approximation. In such a way, the delta function singularity is excluded and the bulk heat transfer is captured for the sake of facilitating the effective training of the neural network. The original problem is then reformulated as a governing correction diffusion or advection-diffusion equation subject to a homogeneous initial condition. The linearly varying thermal conductivity is used to model the soil heterogeneity. We propose a physics-informed neural network to approximate a universal corrector with respect to a single borehole with unit heat extraction rate. As a result, the network is trained by minimizing the physics-informed and data-anchored loss function that is evaluated for sampled conductivity parameters on adaptively selected training points. In addition, we include the location indicator function regarding the source as a feature input of network and find that it helps the network to process the local information. We perform numerical tests to exhibit the effectiveness of the proposed method based on three different analytical models.

AI 总结：研究降落伞悬挂线展开时的力学问题，提出基于物理信息神经网络（PINN）算法预测张力，该算法在计算效率和精度上优于传统方法，还研究了捆绑带参数对线动态张力的调节规律，经对比验证了PINN框架的可靠性和有效性。

链接：https://arxiv.org/abs/2607.12409

机构：Nanjing University of Aeronautics and Astronautics（南京航空航天大学）； Beijing Institute of Space Mechanics and Electricity（北京空间机电研究所）

作者：Xiang Zhao, Ronghui Quan, Yaqi Xiao, Junlin Chen

英文摘要：Parachutes are widely utilized in aviation, aerospace and lifesaving missions. As the initial stage of parachute deployment, suspension line extraction and straightening directly determines the smooth implementation of subsequent inflation procedures. This ultra-short process involves intricate dynamic load variations. Most existing studies adopt numerical integration of ordinary differential equations to calculate line tension, yet this method fails to rapidly acquire tension values at arbitrary positions along suspension lines. This paper develops a physics-informed neural network (PINN) algorithm for tension prediction during line extraction and straightening, which outperforms traditional integration methods in both computational efficiency and numerical accuracy. Furthermore, the regulatory law of binding tape parameters on line dynamic tension is investigated. Comparative validations against flight test data and conventional numerical results verify the reliability and effectiveness of the proposed PINN framework.

AI 总结：研究多变量时间序列异常检测，探索单变量预测基础模型TimesFM的零样本应用于工业MTSAD，评估两种策略，虽未胜过基线，但发现其在捕获时间动态上过于有效致异常难区分，不过在异常边界误差有峰值，对变化点检测有前景。

链接：https://arxiv.org/abs/2607.12454

机构：Salzburg University of Applied Sciences（萨尔茨堡应用科学大学）； Paris Lodron University of Salzburg（萨尔茨堡巴黎洛德龙大学）

作者：Martin Uray, Saverio Messineo, Roland Kwitt, Stefan Huber

英文摘要：Multivariate Time Series Anomaly Detection (MTSAD) is essential for reliability and safety in domains such as industrial process monitoring and financial risk management, yet conventional approaches rely on application-specific models that are costly to train and hard to scale. Foundation Models (FMs), pre-trained on broad data with strong zero-shot generalization, have recently become available for univariate time series forecasting, raising the question of whether they can address MTSAD without task-specific training. We investigate the zero-shot application of a univariate forecasting FM, TimesFM, to industrial MTSAD on the Secure Water Treatment (SWaT) benchmark, evaluating two strategies: treating the FM as a per-feature forecaster with thresholded prediction errors, and as an embedder whose intermediate representations feed standard outlier detectors. Neither of our proposed setups is competitive with established baselines; embeddings reveal only partial separation between normal and anomalous segments, insufficient for reliable detection. The cause is that the FM is too effective at capturing temporal dynamics, yielding low error even within fully anomalous windows, so persistent anomalies become indistinguishable from normal behavior. However, these observations yield valuable insights: the error peaks at anomaly boundaries, indicating FMs reliably detect distribution changes. We conclude that the proposed naive zero-shot FMs are unsuitable for MTSAD but promising for change-point detection.

AI 总结：针对资源受限边缘设备，提出轻量级多尺度自动编码器（LMSAE）网络用于单变量时间序列异常检测。利用离散小波变换提取多尺度特征，采用多尺度损失函数。实验显示其检测性能优、参数少、模型小，适合边缘部署。

链接：https://arxiv.org/abs/2607.12599

机构：Indian Institute of Technology, Delhi（印度理工学院德里分校）

作者：Raheen Junaid Wani, Smruti R. Sarangi

英文摘要：Time-series anomaly detection is increasingly important in IoT systems, sensor networks, and edge monitoring applications, where models must operate under strict constraints on memory, latency, and power consumption. While recent deep-learning approaches have improved detection accuracy, many remain computationally expensive and often fail to capture subtle anomalies due to limited multi-scale sensitivity. Autoencoders are widely used for anomaly detection because they reconstruct normal patterns well, leading to elevated reconstruction errors for anomalous inputs. Their simplicity and efficiency also make them suitable lightweight backbones for handling multi-scale inputs. To address these challenges, we propose a Lightweight MultiScale AutoEncoder (LMSAE) network for univariate time-series anomaly detection, designed to be compact and computationally efficient. LMSAE leverages the Discrete Wavelet Transform (DWT) to extract multi-scale features and employs a multi-scale loss function to improve sensitivity to subtle or hidden anomalies. Experiments on benchmark datasets demonstrate competitive or superior detection performance despite using significantly fewer parameters and a model size of less than 500 KB. LMSAE also achieves low-latency, low-power inference on the NVIDIA Jetson Nano, with 9x reduction in inference latency and 2x reduction in power consumption, making it ideal for edge deployment.

AI 总结：研究智能建筑负荷预测中，推理输入重建后不确定性放置问题。开发统一概率预测框架，比较模块化事后与集成模型内分位数学习方案，用三种深度学习骨干模型实验，发现不确定性放置依赖骨干模型，揭示事后残差分位数局限性。

链接：https://arxiv.org/abs/2607.12730

机构：Istanbul Technical University（伊斯坦布尔技术大学）

作者：Sarah Al-Shareeda, Gulcihan Ozdemir, Heung Seok Jeon

英文摘要： Smart-building load forecasters are often trained offline on dense, multivariate, high-frequency data, but deployment may provide only hourly, feature-limited inputs. Missing features must then be reconstructed, and their errors can propagate through the model. If this input uncertainty is not reflected, prediction intervals may become miscalibrated, affecting demand-response scheduling. Our work examines where uncertainty should be placed once inference inputs are reconstructed. We develop a unified one-day-ahead probabilistic forecasting framework that aligns temporal resolution, reconstructs the unavailable inputs, and derives causal features, and we compare a modular post-hoc residual-quantile scheme with an integrated in-model quantile-learning scheme. The comparison uses three mid-scale Deep Learning (DL) backbones: recurrent, hybrid recurrent, and attention-based Temporal Fusion Transformer (TFT) models, under identical inputs, forecasting horizon, preprocessing rules, and training budgets. Results show that uncertainty placement is backbone-dependent. Integrated quantile learning is most reliable with the TFT, yielding 2.2-3.6% MAPE and 28-83W RMSE on the labeled test window, while producing intervals about 5x narrower than the modular intervals at the closest-to-nominal coverage level. Diebold-Mariano tests support the TFT ranking and the mixed behavior of the recurrent backbones. A reconstruction-sensitivity test shows that reconstructed inputs increase the Quantile Score (QS) by 106% while interval width remains nearly unchanged, indicating that the model does not automatically absorb reconstruction-induced uncertainty. Robustness checks against non-DL baselines and seasonal hold-out weeks support this ranking. Our results expose the limits of post-hoc residual quantiles when inference depends on reconstructed inputs.

AI 总结：研究时间序列预测中上下文的作用，指出频谱指标与添加上下文等因素回答的问题不同，通过替代对等给出覆盖不足诊断，在七个基准测试中验证，表明窗口键控检索等情况，为部署决策提供区分、比较及诊断。

链接：https://arxiv.org/abs/2607.13006

作者：Mert Onur Cakiroglu, Mehmet Dalkilic, Hasan Kurban

英文摘要：A growing family of indices scores how predictable a series is from its spectrum. Practitioners increasingly read these scores as answering a different question: whether \emph{adding context}, a longer lookback, a retrieval plug-in, or a pretrained model, will help. These are not the same question. The value of context is a property of the operating point, not of the series. Any index built from the power spectrum is invariant under phase randomization, whereas the beyond-second-order value that retrieval and foundation models supply is not, because a phase-randomized series is asymptotically Gaussian. We state this as an impossibility result and isolate it with surrogate pairs that fix the spectrum and the marginal by construction. We then give a label-free, configuration-level diagnostic, the coverage deficit, whose principal term measures beyond-spectrum structure as the gain of analog over linear prediction. On seven benchmarks the prediction holds: window-keyed retrieval's value collapses across surrogate pairs (ECL median $+33\%\!\to\!-35\%$, $p{

AI 总结：研究针对线性注意力在状态跟踪和长上下文记忆的局限，提出半直接傅里叶增量注意力（SFDA），通过构造性分块-WY分解，实现精确仿射分块转移等，经数值验证和实验表明其能学习循环记忆，优于禁用相位的KDA基线。

链接：https://arxiv.org/abs/2607.11897

机构：Columbia University（哥伦比亚大学）

作者：Tiantian Zhang

英文摘要：Linear attention replaces softmax attention's growing KV cache with a fixed recurrent state, but this compression limits exact state tracking and long-context memory. We introduce \emph{Semidirect Fourier Delta Attention} (SFDA), a phase-controlled generalization of Kimi Delta Attention that replaces real diagonal decay with block-rotational Fourier control: \[ S_t=(I-\beta_t k_tk_t^*)\Lambda_tS_{t-1}+\beta_tk_tv_t^*, \qquad \Lambda_t=\diag(\alpha_t\odot e^{i\theta_t}). \] Our main result is a constructive chunk-WY factorization for products \(A_t=\Lambda_t-u_tr_t^*\), giving \[ A_t\cdots A_1=\Gamma_t-Y_tM_tW_t^* \] with rank growth bounded inside fixed chunks. This yields an exact affine chunk transfer, formal stability and complexity bounds, and a compact characterization of phase-plus-low-rank memory. We verify the algebra numerically and show in toy state-tracking experiments that SFDA learns cyclic memory where the phase-disabled KDA baseline remains near chance. Fused kernels and large-scale language-model comparisons are left to future work.

AI 总结：针对超维计算中高成本的超向量分解问题，提出量子比特高效框架，引入对数超向量和绑定编码及可逆查找算子，结合改进搜索过程，保持搜索复杂度同时大幅减少量子比特使用，经实验验证效果显著。

链接：https://arxiv.org/abs/2607.11936

机构：University of California, Irvine（加利福尼亚大学欧文分校）

作者：Sanggeon Yun, Hyunwoo Oh, Ryozo Masukawa, Raheeb Hassan, Mohsen Imani

英文摘要：Hyperdimensional Computing (HDC) represents symbols using high-dimensional hypervectors of dimension $D$. In hypervector decomposition, the objective is to recover $F$ constituent hypervectors, each drawn from a codebook of size $N$, from a bound target hypervector. This requires searching over $N^F$ candidate tuples, making the task computationally prohibitive at scale. Recent quantum approach provides a quadratic search advantage, but typically rely on qubit-inefficient $O(D)$-qubit hypervector representations. We propose a qubit-efficient quantum framework for HDC decomposition that reduces the representation cost to $O(\log D)$. The framework introduces logarithmic hypervector and binding encodings, together with a reversible hypervector lookup operator for circuit-level manipulation of dense hypervectors. Combined with a modified Dürr-Høyer search procedure, the method preserves $O(\sqrt{N^F})$ search complexity while substantially reducing qubit usage. Experimental results validate correct similarity computation, accurate decomposition in executable regimes, and significantly improved qubit scaling over baselines based on explicit $D$-qubit hypervector encodings, achieving up to $2{,}000\times$ fewer qubits.

AI 总结：研究提出用可行路径熵衡量智能系统，恢复其理论框架。在GSM8K实验中，增加Qwen2.5-Instruct模型令牌预算，提升了已验证延续能力等指标，表明镜像视界取决于此而非参数数量，支持镜像理论从度量层面解释智能系统能力。

链接：https://arxiv.org/abs/2607.11937

机构： Columbia University（哥伦比亚大学）

作者：Tiantian Zhang (Crystal)

英文摘要：Mirror Theory proposes that an intelligent system should be studied not only by what it represents, but by what coherent continuations it can sustain under repeated reflection. We make this claim operational through \emph{viable path entropy} (VPE), a finite-budget measure of verified continuation capacity. Given a mirror state, a rollout protocol, a verifier, and a mode map, VPE decomposes bounded capability into two parts: the probability of reaching a viable continuation and the diversity of verified continuation modes reached among successful rollouts. This paper restores the full theoretical scaffold behind the measure: intuition as local underdetermining constraint, taste as invariant-selecting pressure, reflection as taste-guided resolution of underdetermination, and geometry as the learned structure that makes future reflection stable. We then instantiate the theory in language-model reasoning experiments on GSM8K. Across Qwen2.5-Instruct models, 32 sampled rollouts per problem, and two reflection horizons, increasing the token budget from 96 to 160 substantially expands verified reachability, reduces zero-reachability, increases verified-mode entropy, and improves smoothed VPE. At 160 tokens, Qwen2.5-1.5B realizes the strongest mirror horizon among the tested models, even though Qwen2.5-3B has more parameters. This shows that mirror horizon is not parameter count, but accessible verified continuation capacity under a bounded reflection protocol. The result supports Mirror Theory as a measure-level account: capability is the structure of viable continuations made reachable, not merely one-shot accuracy or pass@k.

AI 总结：研究在稀缺脑电数据上比较具有RG归纳偏差的RG流变压器和普通变压器，通过睡眠脑电语料库实验发现二者在睡眠分期准确率上无显著差异，但RG流变压器具有可解释性，能在样本外恢复连续谱指数。

链接：https://arxiv.org/abs/2607.11950

机构：Mindverse Computing LLC（Mindverse计算有限责任公司）

作者：Dibakar Sigdel

英文摘要：Brain field potentials are scale-free: their power spectra follow a $1/f^{\beta}$ law whose aperiodic exponent $\beta$ tracks cortical state, and sleep depth in particular is a shift in $\beta$. We ask whether a transformer endowed with an explicit renormalization-group (RG) inductive bias -- the RG-Flow Transformer, which couples ordinary self-attention to a scale-aware stream with a learnable anomalous dimension $\gamma$, block-spin coarse-graining, and an entropy-gated synchronization bridge -- has an advantage over a parameter-matched vanilla transformer on \emph{real, scarce} EEG. Using the PhysioNet Sleep-EDF corpus with a strict leakage-free by-subject hold-out, we (i) benchmark RG-Flow against a param-matched vanilla transformer and a hierarchy-only ablation on 5-class AASM sleep staging, (ii) sweep the per-subject data budget to look for the inductive-bias crossover predicted when data are scarce, and (iii) test whether RG-Flow's learned $\gamma$ tracks the measured spectral exponent $\beta$ out-of-sample -- a quantity the vanilla model does not possess. Across $5$ subjects and $5$ seeds under leave-one-subject-out cross-validation, RG-Flow and the vanilla transformer are statistically indistinguishable on 5-class staging (77.3\% vs 77.0\% accuracy; paired $p=0.294$), and the predicted scarce-data crossover does not appear: vanilla is numerically ahead at every data-limited budget. What does separate the models is interpretability -- RG-Flow recovers the continuous spectral exponent out-of-sample ($\beta$-recovery $R^2 = 0.416$), a capability the vanilla architecture has no analogue for.

AI 总结：研究加权 KNN 回归和软标签预测的数据 Shapley 问题，提出首个伪多项式时间精确算法、认证 FPTAS 等方法，给出复杂度分析和扩展，发布开源库及基准真值，在下游检测中体现精确性价值。

链接：https://arxiv.org/abs/2607.11956

机构：Faculty of Information Technology, Monash University（莫纳什大学信息技术学院）

作者：Zongye Lyu

英文摘要：Data Shapley is the standard principled answer to which training points are worth what, and its k-nearest-neighbor (KNN) specialization is the version deployed in practice: the exact estimator shipped by toolkits such as pyDVL and OpenDataVal. Exact algorithms are known for unweighted KNN and for weighted KNN classification, but weighted KNN regression and soft-label prediction have resisted: the only exact method is an O(N^K) brute force, exponential in neighborhood size K. The obstruction: the weighted regression prediction is a ratio of two coalition-dependent sums, whose normalization denominator breaks the additive, threshold, and duplication structures the prior polynomial algorithms rely on. We close this gap. We give (i) the first pseudo-polynomial-time exact algorithm (polynomial in N and K at fixed lattice precision) for weighted KNN-regression Data Shapley, a counting dynamic program over the joint integer state (sum of w, sum of w*y), verified against exhaustive enumeration with zero mismatch on 12,716 adversarial instances; (ii) a certified FPTAS for continuous weights and targets, with a machine-checkable per-value error certificate never violated across 86,400 checks; (iii) a complexity landscape, including an unconditional Omega(D_w) output-size lower bound and access-model hardness results; and (iv) a weighted soft-label multi-class extension. We release an open-source, CPU-only library and the first exact weighted-regression Data Shapley ground truth. On downstream mislabel detection our exact values are statistically equivalent to Monte-Carlo Data Shapley (dataset-level TOST, n=8, p<10^-4), the pre-registered outcome; the value of exactness is instead determinism, a certified error bound, and an exact reference for auditing estimators: Monte-Carlo did not reproduce the exact top-10% ranking at any budget tested, up to 3,000 permutations (~1.28e6 utility evaluations).

AI 总结：研究针对独立训练神经网络因无共享参考框架及神经坍缩带来的坐标自由度问题，通过五个在MNIST上重建神经坍缩的网络，应用仿射校正对齐，验证了供体特异性功能指纹在特定测试下的可检测性。

链接：https://arxiv.org/abs/2607.11967

机构：H&K Research Studio, Clevix LLC（H&K研究工作室，Clevix有限责任公司）； Banking Academy of Vietnam（越南银行学院）

作者：Truong Xuan Khanh, Phan Thanh Duc

英文摘要： Independently trained neural networks have no shared neuron-index reference frame, so comparing them requires accounting for coordinate freedom. Neural Collapse sharpens this problem: networks converge toward a shared, low-dimensional geometry, raising the question of whether trajectory-specific functional variation remains distinguishable after convergence. We distinguish three claims - detectability, transplantability, and causal persistence - and address the first. Using five independently trained networks reconstructing Neural Collapse on MNIST, we apply a verified affine-correct alignment mapping donor heads into recipient coordinates. Donor-specific functional fingerprints remain distinguishable after recipient-level baseline correction: all 20 ordered donor-recipient pairs are correctly identified, with an exact permutation p=0.0083, robust to a leakage audit. These findings establish detectability under the test used here, but not transplantability or causal persistence. The study shows how alignment, ambiguity diagnostics, and leakage control combine to test cross-network variation in a controlled setting; whether this generalizes beyond it is open.

AI 总结：研究多用户多输入单输出系统中导频到波束成形器设计，提出将ICL-Transformer与导频及波束成形器EDN集成的框架，通过课程学习、自进化机制和失配感知扩展提高性能，优于现有波束成形方案。

链接：https://arxiv.org/abs/2607.11970

机构：Columbia University（哥伦比亚大学）

作者：Yubo Zhang, Xiaodong Wang

英文摘要：We develop an enhanced in-context learning (ICL) framework to improve the performance of pilot-based beamforming in multi-user multiple-input single-output (MU-MISO) systems. The proposed scheme integrates the ICL-Transformer backbone with the pilot encoder-decoder network (EDN) and the beamformer EDN. A crucial feature of our ICL network is that it can handle multiple channel models without retraining, enabled by the construction of model-specific context datasets. To improve convergence and robustness, we introduce three key innovations: (a) a curriculum learning (CL) strategy that smoothly transitions from supervised LMMSE-labeled imitation to unsupervised sum-rate maximization, (b) a self-evolving mechanism that dynamically expands and refines the context datasets for all channel models during CL-based training, and (c) a mismatch-aware extension that incorporates several mismatches into the general ICL framework and bypasses explicit channel calibrations. Ablation studies validate the effectiveness of the in-context architecture and enhanced training strategies. Simulation results over diverse communication environments show that the proposed scheme is able to rapidly adapt to both seen and unseen channel models without gradient-based parameter updates, and can mitigate the mismatch issues via intelligent context constructions. Furthermore, our scheme consistently outperforms the existing beamforming schemes under pilot-based settings, including the WMMSE benchmark and the recent Transformer-based methods.

AI 总结：研究针对精确分子设计，提出JoPMol模型，整合基因表达谱、分子结构信息和化学性质，在统一框架下联合控制生成与优化候选分子，实验显示其性能优于现有方法且泛化能力强，为精确分子设计提供有效模型。

链接：https://arxiv.org/abs/2607.11978

机构：South China Normal University（华南师范大学）； The University of Osaka（大阪大学）

作者：Hang Yuan, Chen Li, Wenjun Ma, Tadahiko Murata, Yuncheng Jiang

英文摘要：Precision molecular design aims to discover personalized drug candidates through joint control of multiple conditions, such as biological relevance and molecular design strategies. Biological relevance reflects cellular functional states under disease or perturbation conditions, while molecular design strategies provide complementary guidance in terms of structural intentions and property optimization. In this study, we propose JoPMol, a jointly controlled precision molecular generative model that integrates biological states encoded by gene expression profiles with molecular structure information expressed in text, and chemical properties quantified by numerical values within a unified modeling framework. This formulation enables coordinated generation and optimization of candidate molecules under joint condition control. Experimental results show that JoPMol outperforms state-of-the-art methods across multiple evaluation metrics. Moreover, JoPMol demonstrates strong generalization ability in both transfer tasks and biologically grounded simulation scenarios, validating its effectiveness for precision molecular design. The source code is publicly available at this https URL.

AI 总结：研究领域专家训练时长对大语言模型模型合并质量的影响，在多领域多模型规模上微调专家模型，保存不同训练步数的检查点并评估五种合并方法，发现方法依赖模式，表明训练时长和合并方法应联合选择。

链接：https://arxiv.org/abs/2607.11997

作者：Nikita Kozodoi, Zainab Afolabi, Jack Butler

英文摘要：Multi-task model merging combines separately trained expert models into a single model that handles all tasks without co-training. Standard practice merges experts at their optimal validation loss. We challenge this convention by systematically studying how training duration of domain experts affects the quality of the merged model. We fine-tune experts on five domains (Math, Code, Instruction Following, Multilingual, and Safety) across three model sizes (Qwen 3.5 0.8B, 2B, and 4B), saving checkpoints from 25\% to 500\% of the optimal training steps and evaluating five merging methods at each duration. Our findings reveal a striking method-dependent pattern: simple averaging degrades sharply with overfitting, while sparsification-based methods achieve their best performance well past the validation optimum. We formalize this through bias-variance decomposition analysis, drawing a parallel to random forests where averaging benefits from high-variance individual learners. These results suggest that training duration and merging method should be chosen jointly rather than independently.

AI 总结：研究聚焦于稀疏自编码器特征，通过重现相关实验发现现有评估混淆不同主张。提出sae - 因果审计工具，发现大量特征存在因果惰性，重新审计细化惰性分类，应用该工具于生产SAE揭示原子碰撞信号。

链接：https://arxiv.org/abs/2607.12166

作者：Mohamed Abdessalem Bal

英文摘要： Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predominantly on correlational recovery metrics -- cosine similarity between ground-truth directions and decoder atoms. We show this conflates two distinct claims: decoder-geometry alignment and encoder-activation behavior. We reproduce the superposition phase diagram of Elhage et al. (2022), identifying a convergence artifact at high sparsity and an under-described diffuse sharing regime at extreme overcompleteness. We reproduce the TopK-versus-L1 comparison of Gao et al. (2024), with direct evidence of L1 shrinkage. Our central result is causal: subjecting every recovered feature to ablation and steering, we find up to 77% of features passing a recovery bar (cosine >= 0.90) in a degraded SAE -- and 9% in a well-trained one -- are causally inert: the matched atom never fires when the feature is present, including matches at cosine ~1.000. We package the method as sae-causal-audit, a model-agnostic instrument with a deterministic pipeline. Re-auditing refines the finding: inertness decomposes by cause into structural inertness (antipodal-pair geometry, present in good SAEs) and competitive inertness (a TopK pathology of degraded SAEs), and by direction into read- and write-inertness, which five antipodal pairs dissociate completely -- unmonitorable yet steerable through the same atom, with steering specificities of 143-310 attached to zero ablation effects. We document why byte-exact reproducibility is unavailable by construction, and propose reporting it as a stack of claims with explicit scopes. Applying the instrument to a production SAE reproduces the pattern at small scale (14% inert) and surfaces an atom-collision signal: a handful of atoms recur as the nearest match for dozens of unrelated concepts, replicated across three batches.

AI 总结：研究提出支持向量注意力（SV - 注意力），通过最大间隔记忆及相关技术实现认证选择和精确遗忘。实验表明其在多种任务中有良好表现，如在真实流数据上的召回率提升，在enwik8数据集上有配对改进，展示了可遗忘检索记忆等能力。

链接：https://arxiv.org/abs/2607.12204

机构：Vy Labs, Inc.（Vy实验室公司）

作者：Vishwajith Ramesh

英文摘要：Attention can be viewed as an online learner over context, yet existing test-time memories cannot certify that dropping a token leaves outputs unchanged or delete its influence outright. We introduce Support Vector Attention (SV-Attention), a max-margin memory whose weights are support coefficients of a one-class SVM with fixed box parameter C. Its active-set partition gives reserve tokens exactly zero weight, certifying output-preserving eviction; a reversible incremental solver deletes a token to recover the state produced by retraining without it under the same C. In fp64 experiments, decrement and refit recover identical partitions whenever the optimum is unique, and their decision functions match to a median deviation of about 10^-9 (10^-13 on learned keys); the 10^-2 worst case is confined to ill-conditioned duplicates and remains below coefficient decay in every regime. The exact path reuses the maintained KKT inverse in a custom backward. Training uses a separate stabilized batched approximation and does not carry the exact-deletion certificate; it reaches 9,125 tokens/s on a 3.22M-parameter model, while remaining 35.8 times slower than an MPS softmax reference. At matched budgets, certified selection reaches 0.86 vs. 0.32 rare-item recall and retains 0.80 vs. 0.05 deterioration hours on real MIMIC-IV streams. We also demonstrate surgical forgetting, exact editing, patient-record deletion, and a forgettable retrieval memory over real sentence embeddings. On enwik8, the hybrid obtains 2.178 BPC vs. 2.383 for a matched-state sliding-window Transformer across seven seeds (8.6% paired improvement, p=0.001); a three-seed TinyStories result is directionally positive but not significant (p=0.057).

AI 总结：研究旨在加速大语言模型智能体，提出为推测器配备三个在线记忆系统，通过从智能体过去轨迹学习提升预测质量。经六个基准测试，记忆增强推测在动作和观察预测上有显著提升，且无损并能跨模型推广。

链接：https://arxiv.org/abs/2607.12236

作者：Yu Li, Qinyuan Ye, Prafulla Kumar Choubey, Jiaxin Zhang, Chien-Sheng Wu

英文摘要：Speculative execution accelerates LLM agents by using a smaller, cheaper model to predict and pre-launch the next step while the environment is idle. However, existing speculators are stateless and discard all information between tasks, preventing prediction quality from improving with experience. We equip the speculator with three online memory systems that learn from past agent trajectories: a contrastive transition table tracking action-sequence statistics, an episodic memory retrieving contextually similar segments, and a confusion tracker suppressing recurring errors. We evaluate this approach on six benchmarks spanning three speculation types: action prediction, observation prediction, and chained prediction. Memory-augmented speculation yields a 19--39\% relative accuracy improvement on action prediction and up to a $2.5\times$ increase on observation prediction tasks with repetitive action spaces. These gains grow continuously as memory accumulates and generalize across speculator models of varying cost. All speculation is lossless because it runs during idle time at zero added wall-clock cost, and the actor's trajectory is identical to non-speculative execution.

AI 总结：研究针对不同状态空间区域动态不同时EDMD效率低的问题，提出聚类加权EDMD，通过期望最大化目标联合学习相空间划分与聚类EDMD算子，在多个系统预测中显著提升效果，降低误差。

链接：https://arxiv.org/abs/2607.12243

机构：AE Studio（AE工作室）

作者：Lorenzo Tomaz, Judd Rosenblatt, Flavio Kicis, Thomas B. Jones, Diogo Schwerz de Lucena

英文摘要：Extended Dynamic Mode Decomposition (EDMD) approximates Koopman operators from data, but a single global operator is inefficient when different state-space regions exhibit distinct local dynamics. We introduce Cluster-Weighted EDMD (CW-EDMD), which jointly learns a soft phase-space partition and a per-cluster EDMD operator. Its Expectation-Maximization (EM) objective assigns each transition based on both geometric proximity and prediction residuals, so clusters specialize where local Koopman models are accurate rather than where the data are dense. On Lorenz, damped pendulum, and Duffing systems, across 36 configurations and 10 seeds, CW-EDMD improves matched-degree EDMD in one-step and 5s-rollout prediction. Across 288 paired comparisons, there are significant error reductions in 258 cases, increases in 4, and no differences in 26. Median one-step error reductions are 57x, 2.7x, and 12x on pendulum, Duffing, and Lorenz, respectively.

AI 总结：针对爱彼迎冷启动个性化推荐难题，引入基于地理IP数据和自适应聚类算法的邻近特征系统，无需个人标识符，在保护隐私的同时实现聚合用户信号生成，经实验验证能显著提升预订量。

链接：https://arxiv.org/abs/2607.12246

机构：Airbnb Inc.（爱彼迎公司）

作者：Wei Jiang, Bin Xu, Hui Gao, Bharathi Thangamani, Weiwei Guo, Sundar Srinivasavaradhan, Tracy Yu, Huiji Gao, Michael Kinoti

英文摘要：Personalization in two-sided marketplaces relies heavily on user-level features, yet for platforms with infrequent, high-consideration purchases, a large fraction of users lack sufficient history for effective recommendation, spanning both paid and organic channels. At Airbnb, a substantial share of search requests comes from logged-out or first-time users, with this challenge especially pronounced on paid-channel landing pages, leaving traditional user-level features unavailable for a large fraction of traffic. Privacy regulations and increasing restrictions on third-party cookies further limit identifier-based tracking for non-essential use cases. This paper introduces Proximity Features, a privacy-compliant feature system that groups users by geographic proximity using geo-IP data and an adaptive clustering algorithm, producing aggregated user-level signals for groups of approximately 1,000 nearby users without requiring a persistent individual identifier at inference time. Privacy is preserved by design: the pipeline operates on consented, aggregated data only within consent-gated privacy controls. The system is deployed in production at Airbnb, serving multiple surfaces including marketing landing pages and destination recommendation, with engagement emails integration under way. Online A/B experiments demonstrate statistically significant lifts in bookings, with the largest gains observed among users with absent or stale history.

AI 总结：研究混合精度量化中如何确定模型高精度部分，通过分析量化损失变化，发现大部分方差由每层效应解释，提出覆盖模型，该模型支持两个预测器，加法模型是最优一阶预测器，覆盖模型本身也有优势，在特定参数模型上实现低KL散度且能解决相关任务。

链接：https://arxiv.org/abs/2607.12266

机构：Baseten Labs, Inc.（巴斯特恩实验室公司）

作者：Joshua Hill

英文摘要：Mixed-precision quantization must decide which parts of a model to keep at higher precision. A common premise, shared by sensitivity-based methods such as HAWQ and CoopQ, is that the loss from quantizing a set of layers can be reconstructed from per-layer or pairwise sensitivities measured in isolation. We test this premise at the 4-bit weight-and-activation precisions now being deployed, treating the change in loss $f(S)$ from quantizing a layer set $S$ as a set function on the Boolean cube and analyzing it through two classical changes of basis. This analysis yields two findings. First, across configurations drawn from the deployment distribution, 85--93\% of the variance of $f$ is explained by per-layer effects alone. Second, a monotone transform of a sum of per-layer terms reproduces $f$'s ranking of configurations, misordering at most 2\% of pairs. We propose the coverage model $f(S)=c\bigl(1-\prod_{i\in S}(1-a_i)\bigr)$, which reproduces the measured variance profile of $f$ to within a few percent from its $L$ fitted break-rates. This structure supports two predictors of a configuration's loss, each with $L+1$ parameters. The additive model is the optimal first-order predictor. By Parseval's identity its mean-squared error equals the variance of $f$ left unexplained by per-layer effects, which we measure on full lattices, estimate out of sample at full-network scale, and report with every result as a certificate of how well any additive model can do. The coverage model itself is the second predictor. As allocators at matched memory, they attain the lowest KL divergence among the compared allocators on models from 30B to 355B parameters. Below four bits, the resulting allocations continue to solve code and reasoning tasks at budgets where allocations from gradient sensitivities no longer produce terminating generations.

AI 总结：研究语言智能体多跳推理链变长性能下降问题，提出SLEUTH方法通过结构化认知工作记忆提升推理，经实验优势随难度增加，还发现证据充分性问题及解决办法，表明组织推理方式是扩展多跳推理关键。

链接：https://arxiv.org/abs/2607.12267

作者：Ning Liu

英文摘要：Language agents that interleave reasoning and tool use degrade sharply as reasoning chains lengthen, even when each individual step is easy. We trace this to context dilution: an agent's investigative state (what it has confirmed, what it suspects, and what it still needs) lives only implicitly in a growing context window, where early discoveries are buried under later retrievals. We introduce SLEUTH, which makes this state explicit and actionable through a structured epistemic working memory: the agent maintains Confirmed Facts grounded to sources, Active Hypotheses ranked by evidence, and Open Questions that directly drive its next action. Across five multi-hop benchmarks and five established baselines, SLEUTH's advantage grows with difficulty, from +5 points on HotpotQA to +11 on 4-hop chains, surpassing Reflexion without multiple episodes. Analyzing where the remaining gap lies, we identify the evidence sufficiency problem: agents often find the answer but fail to commit, exhausting their budget on needless verification. A lightweight commitment trigger fixes this, but only when the agent already maintains structured state: the identical trigger applied to an unstructured agent yields no improvement, isolating organized epistemic state as the necessary condition for effective commitment. Finally, enforcing protocol adherence on a weaker model recovers up to +19 points on the hardest problems, showing that how an agent organizes its reasoning, not raw model capability, is the active ingredient for scaling multi-hop reasoning.

AI 总结：研究无穷小初始化下对角线性网络梯度流动力学，扩展相关定理分析，提出算法1等效表征训练轨迹，证明其收敛到修改后的\(\mathcal{l}_1\)范数最小化问题解，确定隐式偏差并揭示关键几何结构。

链接：https://arxiv.org/abs/2607.12332

作者：Jiajie Zhao, Jianxing Wang, Junjie Yang, Zhiwei Bai, Yaoyu Zhang

英文摘要： We study the gradient flow dynamics of diagonal linear networks for regression tasks under infinitesimal initialization. Extending Theorem 1 from Pesme & Flammarion (2023), we generalize the analysis to both deep diagonal linear networks and a broader class of two-layer diagonal linear networks (as defined in Definition 4.1). Specifically, we demonstrate that the training trajectories of these models can be equivalently characterized by the proposed Algorithm 1. We further prove that this algorithm converges to the solution of a modified $ \mathcal{l}_1 $ norm minimization problem. As a result, we establish that the implicit bias of both network architectures corresponds to a modified $ \mathcal{l}_1 $ norm in the regime of infinitesimal initialization. Additionally, we provide insights into the underlying mechanisms governing these dynamics by identifying the Structural Invariant Manifold (SIM) (Zhao et al., 2026) as the key geometric structure that shapes the learning process.

AI 总结：该研究针对异步时间序列预测问题，提出ReDiTT模型，通过在潜在空间运行并从记忆库检索相似潜在序列作为参考条件，利用交叉注意力机制，实现稳定的长期预测并提高样本多样性，在多个数据集上取得最优性能。

链接：https://arxiv.org/abs/2607.12391

机构：University of British Columbia（英属哥伦比亚大学）； RBC Borealis（加拿大皇家银行北极星）

作者：Saiyue Lyu, Zhitian Zhang, Ruizhi Deng, Thibaut Durand

英文摘要：We present a diffusion based model for asynchronous time series prediction, where the goal is to predict the next inter event time and event type. To address the inherent uncertainty of future events, we introduce ReDiTT, a retrieval augmented conditional diffusion transformer that operates in latent space. ReDiTT retrieves structurally similar latent sequences from a memory bank during both training and inference and incorporates them as reference conditions through cross attention. This retrieval based conditioning allows the model to attend to relevant temporal dynamics and provides global structural guidance for generation. As a result, ReDiTT stabilizes long horizon forecasting and improves sample diversity. Experiments on seven real world datasets demonstrate state of the art performance on next event prediction and long horizon forecasting. Our code is available at this https URL.

AI 总结：研究提出复值玻尔兹曼机PolarBM及其对数尺度变体LogPolarBM，能在极坐标和对数极坐标中处理音频信号，明确建模幅度和相位依赖关系，实验证明其比传统模型有更高建模精度，在多领域有广泛应用潜力。

链接：https://arxiv.org/abs/2607.12417

机构：Graduate School of Informatics and Engineering, the University of Electro-Communications（东京电机大学信息与工程研究生院）； Department of Electrical Engineering and Computer Science, Tokyo University of Agriculture and Technology（东京农业大学电气与计算机科学系）

作者：Toru Nakashika, Kohei Yatabe

英文摘要：Although vast amounts of data, such as audio signal spectra, are naturally represented using complex numbers, conventional machine learning methods often simplify complex-domain problems by employing frameworks designed for real-valued variables. While this simplification offers computational benefits, it discards structural information regarding the inherent relationship between amplitude and phase. In this paper, we propose a novel Boltzmann machine (BM), named PolarBM, capable of naturally handling complex-valued variables in the polar coordinate (i.e., an amplitude-phase representation). PolarBM defines a probability density function for complex variables in which the phase explicitly depends on the amplitude, thereby capturing the physically important relationships of complex-valued signals. Furthermore, to process audio signals in accordance with human auditory perception, we propose LogPolarBM, which models amplitude on a logarithmic scale. This extension yields a flexible conditional probability density function, a power-weighted noncentral complex Gaussian (PW-NCCG) distribution, whose marginal amplitude distribution encompasses the Rice, Nakagami, and noncentral chi distributions as special cases. For practical applications, we also introduce the restricted variants of these proposed models: PolarRBM and LogPolarRBM. Experimental results demonstrate that by explicitly modeling the dependency between amplitude and phase, the proposed RBMs achieve superior modeling accuracy compared to conventional models, including deep neural networks. Although our experiments focus on audio signals, the utility of the proposed BMs is not limited to audio applications; their potential extends widely across various fields of science and engineering that involve complex-valued data, such as wireless communications and quantum mechanics.

AI 总结：研究标签噪声下深度网络记忆损坏标签过程中Fisher秩膨胀现象，推导相关公式并确定条件，通过实验验证其与记忆的关系及随损坏严重程度的变化，建立其作为连接多种因素的谱特征。

链接：https://arxiv.org/abs/2607.12438

机构：University of Southern California（南加州大学）

作者：Satwik Bathula, Anand A. Joshi

英文摘要：Deep networks trained with label noise often learn clean structure before memorizing corrupted labels. We show that this transition leaves a spectral signature in the centered scatter of per-example last-layer gradients. Its effective rank transiently expands during memorization and contracts after corrupted labels are fit. We call this phenomenon Fisher Rank Inflation. Corrupted labels increase effective rank by injecting spectral mass into low-energy or previously unused eigendirections, increasing the entropy of the gradient spectrum. We derive a first-order leave-one-out attribution formula, identify conditions under which corrupted examples contribute more strongly than clean examples, and explain why attribution signals weaken once the normalized Fisher-gradient spectrum stabilizes. We test these predictions on CIFAR-10, CIFAR-100, and CIFAR-10N using SmallCNN, ResNet18, and Vision Transformers. Across settings, Fisher effective rank exhibits a consistent inflation--collapse trajectory aligned with memorization. At peak-rank checkpoints, corrupted examples are enriched among the highest rank-contributing samples, with top-100 noisy fractions from \(69.2\%\) to \(96.2\%\) across five-seed synthetic-corruption experiments and \(94.4\%\pm1.9\%\) on CIFAR-10N. First-order spectral attribution closely matches exact leave-one-out contributions in convolutional models and remains enriched in the Vision Transformer. Peak effective rank increases monotonically with corruption severity, from \(28.88\pm1.95\) under clean training to \(97.09\pm1.78\) at \(60\%\) corruption. In several settings, the retrospectively identified onset of rank inflation precedes observable test degradation. These results establish Fisher Rank Inflation as a spectral signature connecting corrupted-example enrichment, corruption severity, and the transition from structure learning to memorization.

AI 总结：研究大语言模型中置信度的计算基础，利用统计决策置信度框架，通过答案 - 对数差异测试其预测特征，结果表明在多种任务中答案对数可作潜在决策变量读出，为多模态语言模型置信度提供解释并统一研究框架。

链接：https://arxiv.org/abs/2607.12447

机构：Google DeepMind（谷歌深度思维）； École Polytechnique（巴黎综合理工学院）； Princeton University（普林斯顿大学）

作者：Dharshan Kumaran, Viorica Patraucean, Maks Ovsanikov, Petar Veličković, Nathaniel Daw

英文摘要：Reliable confidence -- the probability that a model's own answer is correct -- is essential for the trustworthy deployment of language models. Existing work has largely evaluated confidence by how well it predicts correctness and whether it is calibrated, leaving open a more fundamental question: what does the confidence signal itself represent? Answer logits may reflect a latent decision variable sufficient to compute normative confidence, or instead a heuristic preference signal that combines the available evidence in a non-Bayesian manner. We address this using statistical decision confidence (SDC), a normative framework from computational neuroscience. Treating the answer-logit difference (LD) as a candidate readout of the latent decision variable, we test the qualitative signatures predicted by SDC. Across three perceptual discrimination tasks and a memory-based decision task, spanning three multimodal non-reasoning models and one reasoning model, LD satisfied these signatures -- including the diagnostic correct/error folded-X pattern -- showing that, in these settings, answer logits behave as monotonic readouts of a latent decision variable rather than heuristic preference scores. In complex visual reasoning, LD continued to predict correctness beyond objective task difficulty, but the full geometric signatures of SDC were absent, illustrating the current boundary of the framework when explicit normative process models are unavailable. These results provide a computational account of confidence in multimodal language models, delineate when answer logits behave as readouts of a latent decision variable, and establish SDC as a unifying framework for studying confidence across biological and artificial intelligence.

AI 总结：研究前馈学习中善度衡量问题，基于显式生成模型指出平方善度是似然比检验充分统计量，FF阈值是其边界，还阐述了该方法在不同总体及层间归一化方面的特性及优势。

链接：https://arxiv.org/abs/2607.12501

机构：Sapienza University of Rome（罗马第一大学）

作者：Paolo Giannitrapani

英文摘要：The Forward-Forward (FF) algorithm trains each layer locally, so that a scalar goodness - the sum of squared activations - is high on real inputs and low on contrastive ones, with activations normalized between layers. Both choices are usually treated as heuristics. Under an explicit generative model they are not: the squared goodness is the sufficient statistic of a likelihood-ratio test between two zero-mean populations differing in scale, and the FF threshold is its boundary. It generalizes: anisotropic populations yield a Mahalanobis goodness, the plain square being its isotropic case; heavy-tailed populations yield a saturating statistic whose slope is a posterior precision - divisive normalization - with bounded evidence and an advantage only under aggregation. The same lens characterizes the inter-layer normalization: it must remove the length while preserving per-coordinate energy, explaining a depth collapse we observe under unit-norm normalization; and the pairwise objective admits a scale-inflation shortcut that a whitened goodness removes.

AI 总结：提出基于李群离散强迫欧拉 - 拉格朗日方程的架构，仅用位置数据学习机械系统动力学，尊重系统几何结构与守恒定律，适用于速度数据不佳情况，可扩展到多体系统，在合成与真实数据集上表现良好。

链接：https://arxiv.org/abs/2607.12627

机构：SINTEF Digital（挪威科技工业研究院数字技术部）； Instituto de Ciencias Matematicas (CSIC)（西班牙国家研究委员会数学科学研究所）

作者：Martine Dyring Hansen, Marta Ghirardelli, Elena Celledoni, David Martin de Diego, Brynjulf Owren

英文摘要：We propose an architecture for learning the dynamics of mechanical systems based on discrete forced Euler-Lagrange equations on Lie groups using only position data. By formulating the dynamics directly on manifold-valued configuration spaces, the method naturally respects the geometric structure of the systems and preserves geometric invariants and conservation laws. The reliance on position measurements alone makes the framework applicable in settings where velocity data are unavailable or noisy. The approach extends naturally to multibody systems, accommodates external control inputs, and demonstrates strong performance on both synthetic and real-world datasets.

AI 总结：研究针对纵向电子健康记录生成中标准模型的不足，提出AdaPCLA框架，通过数据分布感知训练策略、模拟退火训练及零样本分布控制实现自适应拟合与生成，理论分析刻画相关界限，实验验证其在多方面有提升。

链接：https://arxiv.org/abs/2607.12645

机构：University of Science and Technology of China（中国科学技术大学）； Sun Yat-sen University（中山大学）

作者：Shuai Cui, Chen Wenxuan, Wenjie Du, Jian Lou, Dan Li, Wenjie Feng

英文摘要： Generative modeling of longitudinal Electronic Health Records is increasingly important for privacy-preserving research, yet standard autoregressive models tend to underrepresent the co-occurrence structure of tail events (i.e., diseases, symptoms), reducing the fidelity and faithfulness of generated data for rare subpopulations. To this end, we propose AdaPCLA framework, which enables generative models to adaptively fit and generate EHR data through a data distribution-aware training strategy; this is achieved by internalizing data knowledge parameters by simulated annealing training. It also supports training-free adaptation to a diverse clinical population for generation through zero-shot distribution control. Moreover, our theoretical analysis characterizes rare-code logit updates through the label-wise empirical NTK and derives a prior-internalization bound for how annealing speed and NTK conditioning affect retained prior signals. Experiments on real-world data show that AdaPCLA achieves consistent gains in tail plausibility, downstream utility, and zero-shot control; in particular, it improves TailPairSeen over HALO by 114.2% on MIMIC-III and 65.1% on MIMIC-IV, outperforms GPT-style generation by 3.5% F1 for zero-shot cross-population adaptation.

AI 总结：研究大语言模型可提取记忆的有效性问题，提出通过匹配比较测量训练与非训练序列生成概率来判断记忆，形式化共形测试和普查两种匹配比较方式并揭示问题，完善可提取记忆定义，给出有效声明及现实预算下的生成要求。

链接：https://arxiv.org/abs/2607.12649

机构：Yale（耶鲁大学）； AVERI（AVERI（未提及常见中文名，可直译为“应用价值与伦理研究所”之类，具体需结合实际背景））； Stanford（斯坦福大学）； Google Research（谷歌研究院）； Google DeepMind（谷歌深度思维公司）； Cornell（康奈尔大学）

作者：A. Feder Cooper, Marika Swanberg, Jamie Hayes, Lea Duesterwald, Christopher De Sa, Daniel E. Ho, Mark A. Lemley, Percy Liang

英文摘要：Recent work on extractable memorization in LLMs suffers from two contrasting validity problems. Some studies overstate extraction, e.g., relying on sequences too short to distinguish memorization from predictability. Others imply that extraction is unreliable evidence of memorization, since models can also reproduce real-world text they weren't explicitly trained on. In different ways, both overlook what makes a valid extraction claim: the model must generate a training sequence with high enough probability to indicate memorization. To determine what's high enough, one has to perform a matched comparison: measuring the generation probabilities of both the training sequences of interest and comparable non-training sequences. Because non-training sequences cannot have been memorized, their probabilities provide a baseline for predictability; a training sequence exceeding this baseline provides evidence of memorization. We formalize matched comparisons in two ways: (1) a conformal test that calibrates a threshold to a chosen FPR when training and non-training sequences are sampled from populations, and (2) a census that calibrates against a matched non-training document when the object is a single document (e.g., a book). We show that matched comparisons enable rigorous, calibrated memorization claims, and reveal where prior setups have validity issues. For instance, on Wikipedia OLMo 2 32B reproduces non-training 10-token suffixes roughly 24% as often as training ones: that share of the training generation rate reflects false positives, not memorization. For Llama 3.1 70B on books, the thresholds we calibrate are as low as 1e-27, supporting memorization claims for sequences that no feasible sampling budget would extract. Based on these results, we refine "extractable memorization" to require a valid memorization claim and near-certain generation within a realistic budget.

AI 总结：研究旨在消除语言模型经验推理中的幻觉，提出基于Lean 4的EG-VAR工具调用架构，通过工具验证公理等生成可验证声明，经实验在数值推理等测试中表现良好，定位为高风险经验声明的技术治理接口，可审计相关条件并转化错误为审计目标。

链接：https://arxiv.org/abs/2607.12650

作者：Junyu Ren

英文摘要：Tool access alone does not make LLM empirical reasoning governable: accepted outputs need not descend from attested evidence, and accepted deductions need not hold up under formal scrutiny. We present EG-VAR (Evidence-Grounded Verified Agentic Reasoning), a Lean 4-based tool-calling architecture in which the Lean kernel is the sole minter of Verified claims via tool-attestation axioms and declared source lifts. Every verified output structurally descends from an attested tool call (Thm. 3.1) and a kernel-checked chain of valid inference (Thm. 3.2); residual outputs are honest Abstain with a replayable audit trail. On a subcollection of TableBench numerical reasoning (n=120), EG-VAR attains 120/120 versus a 95% same-tool baseline; on counterfactual stress tests (5 domains x 2 models), EG-VAR stays 100% source-faithful while same-tool drops to 80-90% (no-tool 50-80%). With the LLM as deployment-time formalizer, residual semantic-formalization error is 3.3% on Sonnet and 1.7% on Opus. We position EG-VAR as a technical-governance interface for high-stakes empirical claims: a formal sidecar makes the target proposition, source scope, evidence boundary, proof obligation, and abstention condition auditable, eliminating unsupported Verified outputs today while turning formalization errors, lift and source-authority disputes, ambiguities, and abstentions into explicit audit targets. Over time, typed sidecars in datasets, APIs, public records, and AI-generated documents can amortize this formalization burden into reusable infrastructure.

AI 总结：研究针对Transformer模型注意力机制计算瓶颈，提出自适应向量量化（AVQ）注意力，基于注意力重要性分配码本容量，前向传播识别重要代码并用子码字细化，开发相关实现，保持\(\mathcal{O}(MN)\)复杂度，提升精度-效率权衡。

链接：https://arxiv.org/abs/2607.12789

机构：QUVA Lab, University of Amsterdam（QUVA实验室，阿姆斯特丹大学）； AMLab, Informatics Institute, University of Amsterdam（AMLab，阿姆斯特丹大学信息学院）； AI4Science Lab, University of Amsterdam（AI4Science实验室，阿姆斯特丹大学）； Korteweg-de Vries Institute for Mathematics, University of Amsterdam（科特韦格 - 德弗里斯数学研究所，阿姆斯特丹大学）； Qualcomm AI Research（高通人工智能研究公司）； FunAI Lab, University of Technology Nuremberg（FunAI实验室，纽伦堡工业大学）

作者：Winfried van den dool, Patrick Forré, Amir Habibian, Yuki M. Asano, Max Welling

英文摘要：The $\mathcal{O}(N^2)$ complexity of attention over $N$ tokens remains a computational bottleneck in transformer models. Vector-Quantized (VQ) attention reduces this to $\mathcal{O}(MN)$ by representing keys with $M$ codewords, but applies uniform codebook capacity regardless of where attention mass concentrates: high-attention regions of key space may be coarsely approximated while low-attention regions waste representational capacity. We propose Adaptive Vector-Quantized (AVQ) Attention, which adaptively allocates codebook capacity based on attention importance. Starting from a small set of codewords, our method identifies the most important codes during the forward pass and refines them with pre-learned child codewords, achieving fine-grained quantization where it matters most while maintaining coarse quantization elsewhere. We develop an implementation using custom Triton kernels that enables the full adaptive refinement process, including importance scoring, child codeword insertion, and parent contribution replacement, to be carried out within the tiled computation paradigm of Flash Attention with minimal overhead. Our approach maintains $\mathcal{O}(MN)$ complexity while achieving improved accuracy-efficiency trade-offs compared to fixed-codebook VQ-attention.

AI 总结： 针对聚类张拉整体结构找形及物理特性预测难题，提出基于能量的学习框架，将总势能最小化和本构关系纳入训练目标以同时预测相关物理量，经实验验证该方法有潜力且能准确预测结构特性。

链接：https://arxiv.org/abs/2607.12888

机构：University of Kentucky（肯塔基大学）； University of Houston（休斯顿大学）

作者：Jing Qin, Muhao Chen

英文摘要：Tensegrity form-finding and physical property prediction are fundamental inverse problems in structural mechanics, which aim to determine equilibrium configurations and internal force distributions. These problems are challenging due to strong nonlinearity arising from the coupling between geometry and forces, the need to ensure structural stability, and the enforcement of constraints such as boundary conditions and symmetry. Moreover, traditional methods often lack robustness to noise and outliers. This paper proposes an energy-based learning framework for clustered tensegrity form finding and physical property prediction. The proposed approach incorporates total potential energy minimization and constitutive relations into the training objective, enabling the simultaneous prediction of equilibrium nodal configurations and associated physical quantities, including member forces and force densities. By incorporating energy-based physical losses directly into the learning process, the framework improves physical consistency, robustness, and data efficiency. Numerical experiments on tensegrity structures, including prism and lander systems, show the great potential of the proposed approach and demonstrate its capability for scalable form finding and accurate prediction of structural properties.

AI 总结：研究提出CoCo损失函数学习归一化且结构良好的表示，鼓励类内坍缩和类间对比，有更优初始化等优势。实验表明其在表格数据集上性能与现有方法竞争，能促进类聚类和更快收敛，是学习判别性表示的有效目标。

链接：https://arxiv.org/abs/2607.12916

机构：Departamento de Ingeniería Informática, Universidad Autónoma de Madrid（马德里自治大学信息工程系）

作者：Blanca Cano-Camarero, Ángela Fernández-Pascual, José R. Dorronsoro

英文摘要：In this work, we introduce CoCo, a loss function aimed at learning normalized and well-structured representations. The proposed loss encourages intra-class collapse and inter-class contrast while preserving sufficient flexibility for neural networks to approximate geometrically optimal embeddings with large angular separation between classes. We provide a theoretical analysis positioning CoCo with respect to related objectives such as dot regression and cross-entropy, showing that the new proposed loss benefits from closer initialization to the optimal configuration, more informative gradients, and stronger incentives for class-wise representation collapse. Extensive experiments on diverse tabular datasets from the OpenML-CC18 benchmark show that CoCo achieves competitive performance with state-of-the-art methods, including kernel SVM, Random Forest, dot regression, and cross-entropy-based neural networks. In addition, both theoretical arguments and empirical analyses demonstrate that the proposal promotes tighter class clustering and faster convergence. These results highlight CoCo loss as an effective objective for learning discriminative representations while maintaining competitive predictive performance.