机器学习学术速递[7.10]

AI 总结：研究针对标准注意力应用于偏微分方程的局限，提出局部线性Transformer（LLT）。它结合线性全局注意力与局部空间混合并纳入相关信息，在多类偏微分方程问题上评估，结果显示其准确且高效，降低训练时间并能处理复杂数据集。

链接：https://arxiv.org/abs/2607.07718

作者：Oded Ovadia, Eli Turkel

英文摘要：Neural operators have become a common approach for learning PDE solution maps and accelerating numerical simulations. Transformer-based neural operators are of particular interest, since attention can learn long-range dependencies in the computational domain. However, standard attention has two major limitations when applied to PDEs: it scales quadratically with the number of computational nodes, and it lacks an explicit bias toward local interactions. To address these issues, we introduce Local Linear Transformer (LLT) for PDE operator learning. The architecture combines linear global attention with local spatial mixing, and incorporates coordinate and geometry information. We evaluate LLT on several PDE problems, including elasticity, plasticity, airfoil flow, pipe flow, and Darcy flow. The reference data for these problems span finite-element, finite-volume, and finite-difference discretizations on structured and unstructured meshes. Compared with other neural-operator and transformer baselines from prior studies, LLT achieves competitive or lower relative $L_2$ error across these problems. On matched structured discretizations, wall-clock time per training iteration is reduced by factors of 1.8 to 2.5 relative to Transolver. We also scale the approach and apply it to a three-dimensional car aerodynamics dataset with 32,186 unstructured mesh points per sample. Together, these results indicate that LLT provides an accurate and computationally efficient operator for PDE problems across discretizations, mesh types, and problem settings.

AI 总结：研究针对块稀疏注意力截断短视问题，提出信息价值路由器，通过测量 top-k 截断决定性程度调整保留集。该方法与主干无关，可堆叠现有方法。实验表明其在多个模型上提升召回率，保留准确率，且减少运行时间。

链接：https://arxiv.org/abs/2607.07724

作者：Thomas Rossi

英文摘要：Block-sparse attention scales long-context language models by replacing the O(N^2) softmax with a per-query top-k selection over key blocks. This cutoff is myopic: when the k-th and (k+1)-th blocks are nearly tied in score, the selector commits without spending extra budget, and a dropped block carrying answer evidence is unrecoverable downstream. We propose a value-of-information router that measures, for each query, how decisively the top-k cut was made, and doubles the kept set for the queries where that gap is smallest; the rule is backbone-agnostic and stacks with existing block-scoring methods such as Quest. On LongBench-v2 medium at n=215 (the entire dataset subset), router-on-Quest reaches paired recall 0.75 vs. top-k 0.47 -- +28 pp over the SSA-style baseline (McNemar p<0.01) -- and lands within 2 pp of dense on RULER NIAH multikey at the same context. The lift reproduces on four models from three architectures (Qwen2.5, Mistral-Nemo, Qwen3.6). At 128K, the router preserves 0.81 and 0.89 of dense accuracy on Qwen2.5-7B-1M and Qwen3.6 (vs. SSA-style top-k at 0.09 on the former) while the fused selection-plus-kernel pipeline runs at 0.62x and 0.80x dense wall time.

AI 总结：研究针对现代语言模型长上下文应用中零样本上下文扩展问题，提出Jet-Long方法，通过动态双焦点RoPE及相关技术，在推理时开销小，在多种模型和基准测试中表现优异，还能推广到其他架构且超参数弹性强。

链接：https://arxiv.org/abs/2607.07740

作者：Haozhan Tang, Zerui Wang, Yuxian Gu, Song Han, Han Cai

英文摘要： Modern LLMs are increasingly deployed in long-context applications such as retrieval-augmented generation, repository-level coding, and agentic workflows whose accumulated reasoning and tool traces routinely push the input an order of magnitude past the pretraining window, making zero-shot context extension the dominant deployment path for open-weight checkpoints. Most existing zero-shot methods fix a single rescaling factor up front, so an aggressive factor sacrifices short-context fidelity while a conservative one breaks down at long contexts. We propose Jet-Long, a tuning-free zero-shot method that pairs a local RoPE-faithful window with a long-range window whose rescaling factor adapts dynamically to the current sequence length, recovering the base model exactly at short inputs while extrapolating cleanly at long ones. An inclusion-exclusion attention merge and an on-the-fly RoPE correction rotation make the bifocal construction essentially free at inference; fused into a single CuTe kernel, long-context prefill reaches up to $1.39\times$ FA2 throughput on H100 (approaching the Hopper-only FA4), and single-batch generation incurs $\le 4\%$ overhead at every length. On Qwen3-1.7B/4B/8B up to 128K context, Jet-Long leads RULER by $+4.79$/$+2.18$/$+2.03$~pp over the strongest baseline at 1.7B/4B/8B, achieves the best overall accuracy on HELMET-RAG (a benchmark identified by HELMET as the most efficient predictor of downstream long-context performance) and attains the lowest PG-19 perplexity. Jet-Long also generalizes to hybrid attention architectures such as Jet-Nemotron for further long-context improvement without retraining, and remains hyperparameter-resilient for ease of deployment.

AI 总结：研究受生物神经元启发，引入MetaNCA框架学习局部规则自组织神经网络权重。提出新颖权重变换器架构，经训练的规则网络无需反向传播生成多样任务网络，在MNIST和CIFAR - 100上实验，证明其能生成权重并泛化到未见架构。

链接：https://arxiv.org/abs/2607.07743

机构：Mythos Scientific（神话科学）

作者：Meet Barot, Daniel Berenberg, Sina Khajehabdollahi

英文摘要：Self-organization is an emergent property of life, driven by the collective behavior of individual components acting on local information. Biological neurons, through local interactions transmitted through synapses, are able to learn efficiently and can adapt their connections over an organism's lifespan. Motivated by these desirable properties of adaptability and local interaction, neural cellular automata (NCA) models have been successful at learning morphogenesis solely through local update rules, demonstrating stability over many updates and robustness to perturbations. In this work, we introduce Meta Neural Cellular Automata (MetaNCA), a framework that learns local rules which self-organize the weights of artificial neural networks. A learned rule network iteratively updates the weights of a task network using only local interactions on the computation graph. We propose a novel Weight Transformer architecture for the local rule network, which uses linear attention to aggregate signals from neighboring weights and hidden states. Once trained, the rule network generates task networks of diverse architectures without backpropagation. We show that MetaNCA generates weights for feedforward MLPs, CNNs, and ResNets on MNIST and CIFAR-100, scaling to networks of 2 million parameters. We further show that MetaNCA generalizes to architectures not seen during meta-training, and that architectural diversity in the training phase strengthens this generalization.

AI 总结：针对低资源代码生成难题，提出三阶段管道，先将推理计算左移合成训练数据，再微调嵌入句法先验，最后用可验证奖励强化学习，相比现有方法提升性能，减少数据使用和成本，且能推广到新语言。

链接：https://arxiv.org/abs/2607.07748

机构：WSO2

作者：Didula Samaraweera, Anjana Supun, Srinath Perera

英文摘要：Large Language Models achieve strong code generation for high resource languages like Python and Java but suffer sharp performance drops on Low-Resource Programming Languages~(LRPLs) such as Julia. Improving Small Language Models~(SLMs) for these languages faces a trilemma: Supervised Fine-Tuning~(SFT) is bottlenecked by data scarcity, inference-time scaling is too expensive for deployment, and Reinforcement Learning from scratch yields near zero advantages. We propose a three-phase pipeline that resolves this trilemma by decoupling syntax acquisition from algorithmic reasoning. First, we \emph{left-shift} inference-time compute to an offline data synthesis engine that uses iterative compiler and test feedback to generate verified training examples. Second, we fine-tune an SLM on this synthetic, verified data to embed strong syntactic priors. Third, we apply Reinforcement Learning with Verifiable Reward~(RLVR) grounded by language-agnostic Input/Output tests, where the SFT prior constrains exploration away from syntax errors. Applied to Qwen3-8B, our pipeline improves pass@1 by up to +7.6 points on MultiPL-E and +14.2 points on the Agnostics LiveCodeBench for Julia compared to SOTA results. Furthermore, the pipeline only used $\frac{1}{3}$ data and $\frac{1}{6}$ cost over the previous state-of-the-art. We further demonstrate that the pipeline generalizes to Ballerina achieving 49.7\% MultiPL-E Pass@1, a language with near-zero pretraining representation. Ablations confirm that both the SFT phase and execution-grounded rewards are necessary for stable training.

AI 总结：研究提出一种经典 - 量子混合的图像分类策略，量子部分含图像幅度编码等操作，多个专家用不同参数处理图像并提取特征，经典部分联合处理特征进行预测。实验表明该策略优于单个专家分析，降低预测失败率，在GPU上开销适中，还可在量子处理器执行。

链接：https://arxiv.org/abs/2607.07754

机构：Centre for High Energy Physics, Indian Institute of Science（印度科学研究所高能物理中心）

作者：Kumari Jyoti, Rohith Babu, Apoorva D. Patel

英文摘要：Pattern recognition problems arise in a variety of physical image processing situations, and convolutional neural networks are a popular scheme for the required feature extraction and classification tasks. The classical networks use diffusion-based smearing and block-wise pooling to downsample the image data and capture important structural features. In this work, we propose and demonstrate a more efficient quantum-inspired strategy involving a mixture of experts. It is a hybrid classical-quantum framework. The quantum part consists of amplitude encoding of the images, convolution using local unitary operations, multiple experts processing the same image with different parameters, and feature extraction using quantum stabiliser codes. The classical part then jointly processes the features extracted by different experts using a standard fully connected neural network for image class prediction. Using MNIST and Fashion-MNIST datasets as benchmarks, we demonstrate that the joint expert analysis outperforms the individual expert one, as well as reduces the failure rate of image class prediction by around a factor of two. The overhead of our quantum-inspired strategy is only moderate on GPU workstations, which makes our proposal a practical alternative to existing classical schemes. We also point out how the quantum part of our framework can be executed on a quantum processor.

AI 总结：研究世界模型在可变时间分辨率下预测动力学的问题，利用哈密顿生成网络（HGN），指出其在非保守环境中时间泛化失效的问题及原因，通过针对性修复实现稳定动力学预测，推荐连续时间视频生成中时间泛化的策略。

链接：https://arxiv.org/abs/2607.07763

机构： Department of Computer Science, Southern Methodist University（南卫理公会大学计算机科学系）

作者：Eli Laird, Corey Clark

英文摘要：World models are typically trained to predict discrete-time physical dynamics with a fixed step size baked into the model weights, preventing prediction at variable temporal resolutions. This matters for hierarchical planning, sim-to-real transfer, and scientific or game-engine applications that must query the same dynamics at multiple timescales. Hamiltonian Generative Networks (HGN) offer a principled path forward, grounding predictions in a continuous-time energy function that is, in principle, independent of the observation frame rate. In practice, however, their temporal generalization breaks down in non-conservative settings. We show that in externally forced, dissipative environments, HGN rollouts at step sizes beyond the training regime fail due to distinct failure modes, including latent magnitude growth driven by an unconstrained action-force map, and global truncation error accumulation from an under-resolved integrator. We identify a targeted fix for each mechanism and demonstrate stable dynamics prediction at temporal resolutions well outside the training distribution. In a detailed analysis, we recommend several strategies for enabling temporal generalization in continuous-time video generation.

AI 总结：研究深度标量线性网络梯度下降动力学，发现最优深度方向学习率缩放取决于数据，与数据无关的规则不适用，数据依赖最优缩放使学习动力学与数据无关、对深度弱依赖，在含残差连接网络中也有类似效应。

链接：https://arxiv.org/abs/2607.07884

作者：Yedi Zhang, Peter E. Latham, Leena Chennuru Vankadara, Andrew Saxe

英文摘要：In this short note we consider the gradient descent dynamics of deep scalar linear networks, $f(x) = \prod_{l=1}^L w_l x$, which enjoy exact time-course solutions for any integer depth. We show that even in this minimal model, the optimal depth-wise learning rate scaling depends on data, whereas data-agnostic scaling rules fail to transfer across depths. Under the data-dependent optimal scaling, the learning dynamics is independent of data and weakly dependent on depth, resulting in a constant linear convergence rate across all depths including infinity. We further show similar data-dependent effects in deep scalar linear networks with residual connections.

AI 总结：研究对softmax注意力和四种循环线性注意力架构比较，用通用符号表达机制差异。通过训练350M参数模型实验，发现不同架构在损失、吞吐量等方面表现不同，还引入评估跨层路由机制，如CLVR可降低DeltaNet和门控DeltaNet的最终验证损失。

链接：https://arxiv.org/abs/2607.07953

机构：ETH Zurich（苏黎世联邦理工学院）； ETH AI Center, ETH Zurich（苏黎世联邦理工学院ETH人工智能中心）

作者：Tommaso Cerruti, Tim Rieder, George Rowlands, Lingfeng Jin, Imanol Schlag

英文摘要：Self-attention lets each token retrieve information from the full context, but its quadratic cost in sequence length limits training and inference at long context. This paper presents a comparative study of softmax attention and four recent recurrent linear-attention architectures: DeltaNet, Gated DeltaNet, Kimi Delta Attention, and Gated DeltaNet-2. We express these mechanisms in a common recurrent-memory notation, making explicit how they differ in expressivity, memory decay, erase and write control, training throughput, and implementation complexity. Our experiments center on 350M-parameter models trained for 15B tokens, and include optimizer and learning-rate comparisons, hybrid-versus-pure stack comparisons, sequence-length runtime measurements, larger DeltaNet runs at 1.3B and 3B parameters, and a small set of downstream evaluations. The reported speed results measure training throughput and iteration time; we do not provide an empirical inference-speed benchmark. Within the reported 350M-parameter, 15B-token sweep, Kimi Delta Attention with Muon reaches the lowest final validation loss, a pure Gated DeltaNet stack trained with AdamW has the highest normalized training throughput, hybrid stacks generally improve loss at a throughput cost, and Muon consistently lowers final validation loss relative to AdamW in the matched architecture settings we evaluate. We introduce and evaluate lightweight cross-layer routing mechanisms for DeltaNet-style memories. The most natural DeltaNet-inspired formulation, forwarding a lower layer's delta-rule write error into the next layer's value target, does not improve over matched baselines. Routing into the aligned hidden stream and forwarding the write value instead yields a modest improvement in the matched runs we report: Cross-Layer Value Routing (CLVR) lowers final validation loss for both DeltaNet and Gated DeltaNet.

AI 总结：针对神经PDE求解器内存消耗大及单节点瓶颈问题，提出PGD-NO神经算子，通过预计算几何分解、迭代算法提取令牌解耦特征提取与解查询，实现线性内存扩展，在工业基准测试中表现佳且具可解释性，为大规模工业设计应用提供方案。

链接：https://arxiv.org/abs/2607.08025

作者：Weiheng Zhong, Jing Bi, Victor Oancea, Hadi Meidani

英文摘要：While neural PDE solvers have demonstrated significant potential for accelerating engineering simulations, existing architectures remain constrained by high memory consumption and the single node bottleneck, where the maximum processable mesh resolution is strictly limited by the VRAM of a single compute unit. To address these challenges, we propose PGD-NO, a neural operator with Precomputed Geometry Decomposition, that relocates the computational overhead of geometric encoding to a deterministic pre-computation phase. By utilizing an iterative geometry decomposition algorithm to extract geometry tokens, our model decouples feature extraction from solution querying. This architecture enables linear memory scalability, allowing high fidelity learning on meshes exceeding 10 million nodes, a scale where existing architectures typically encounter memory exhaustion. PGD-NO demonstrates competitive predictive accuracy across diverse industrial benchmarks and provides intrinsic interpretability through attention mechanisms. By effectively overcoming traditional mesh-size constraints, PGD-NO offers a robust and efficient solution for the next generation of large-scale, high-fidelity industrial design applications.

AI 总结：针对人工智能两用困境及多模型部署成本高问题，提出梯度路由辅助模块（GRAM）预训练方法，能在推理时消融模块功能。实验显示其可有效禁用目标功能且抗微调恢复，训练成本低，紧密跟踪数据过滤。

链接：https://arxiv.org/abs/2607.08077

作者：Ethan Roland, Murat Cubuktepe, Erick Martinez, Stijn Servaes, Keenan Pepper, Mike Vaiana, Diogo Schwerz de Lucena, Judd Rosenblatt, Addie Foote, Cem Anil, Alex Cloud

英文摘要：AI developers face a dual-use dilemma. An AI capability that helps one user cure a disease can help another synthesize one. This dilemma could be resolved with access control, limiting dual-use AI capabilities to trusted deployments with a legitimate need. A gold standard for access control would be to serve separate models with different capabilities to different users. However, training and deploying multiple models is prohibitively expensive. To address this challenge, we propose gradient-routed auxiliary modules (GRAM), a pre-training method that adds modules to a neural network and selectively updates them to induce specialization. Ablating a module at inference time removes its capability from the network, approximating a model trained on filtered data. We evaluate GRAM on synthetic stories and realistic dual-use data spanning virology, cybersecurity, nuclear physics, and specialized code. These experiments show that GRAM disables targeted capabilities while preserving the rest, and resists their recovery under finetuning better than post-hoc unlearning. Most importantly, a Chinchilla-optimal scaling analysis from 50M to 5B parameters shows that the gap between data-filtered and full-data models widens with scale on removed capabilities but stays small on retained ones, and that GRAM closely tracks data filtering. GRAM's training cost is independent of the number of supported capability profiles, yielding a 5x reduction over data filtering in our 5-profile setting.

AI 总结：研究零样本模型大小插值中如何选择学生层，将其转化为优化问题并证明可看作无环图最短路径问题。通过实验揭示修补对插值行为的影响，介绍KLPatch算法，为模型大小插值提供理解与实用指导。

链接：https://arxiv.org/abs/2607.08170

机构：Harvard University（哈佛大学）； Kempner Institute（肯普纳研究所）； IST Austria（奥地利科学技术研究所）

作者：Sara Kangaslahti, Jonathan Geuter, Nihal V. Nayak, Marco Fumero, Francesco Locatello, David Alvarez-Melis

英文摘要：Zero-shot model size interpolation aims to create new models of intermediate target sizes by combining existing models without additional training. Recent work on boomerang distillation [Kangaslahti et al., 2026] shows that a student language model distilled from a larger teacher can be expanded by iteratively patching its layers, replacing student layers with contiguous blocks of teacher layers to obtain models whose size and performance interpolate between the student and the teacher. In this work, we provide the first systematic study of student-layer selection for model size interpolation. We cast finding the optimal layer subset for each model size as an optimization problem and prove it can be viewed as a shortest-path problem in a certain acyclic graph. In experiments, we show that patching strongly shapes interpolation behavior, with effects that vary substantially across model families. We find that simple sequential strategies--patching either from the first layer to the last or from the last to the first--often achieve surprisingly strong performance in practice. We further introduce KLPatch, a greedy patching algorithm based on KL divergence, which often improves over last-to-first patching and approximately solves the optimization problem. Together, our results provide a principled understanding of how layer patching affects model size interpolation and offer practical guidance for constructing near-optimal interpolated models.

AI 总结：针对推荐系统中估计原始空间条件期望的问题，提出PIT-SUN框架，通过经验边际表定义相关内容，应用乘法SUN恢复估计期望，实验表明该方法在多方面有提升且部署开销小。

链接：https://arxiv.org/abs/2607.08202

作者：Mingyu Zhao, Zhaohan Li, Zhenxiong Miao, Xu Zhang, Dewei Leng, Yanan Niu, Kun Gai

英文摘要：Estimating original-space conditional expectations is central to value-driven recommender systems, including dwell time, GMV, and LTV forecasting. Standard MSE is expectation-consistent in principle, but its gradients become unstable on heavy-tailed, zero-inflated, and multimodal targets, causing mean collapse and tail shrinkage. Target transformation alleviates this scale conflict, yet any useful nonlinear marginal transform loses expectation consistency under direct inversion. This is not an implementation oversight: a direct inverse-transform estimator is universally expectation-consistent only when the inverse transform is affine, which cannot simultaneously provide bounded tail compression. Existing conditionally linear recovery methods restore expectation consistency, but still leave open which coordinate, inverse lookup, recovery base, and deployment monitor should be selected for sparse complex marginals. We propose \textbf{P}robability-\textbf{I}ntegral-\textbf{TranS}formed \textbf{Un}biased recovery (\textbf{PIT-SUN}), a deployable empirical marginal recovery framework. PIT-SUN uses one empirical marginal table to define a bounded normal-score coordinate, its inverse-quantile lookup, a variance-controlled recovery base, and drift monitoring, then applies multiplicative SUN recovery to estimate the original-space expectation instead of directly inverting transformed predictions. Experiments on synthetic distributions, public benchmarks, large-scale industrial datasets, and online deployment show robust improvements in point accuracy, calibration, and ranking quality with lightweight deployment overhead.

AI 总结：针对现实时间序列复杂动态及现有预测架构问题，提出RhyMix混合神经架构，通过并行双路径建模及自适应门控机制集成互补分支，在多数据集测试中性能领先，且轻量级、低延迟，适用于边缘设备和实时部署。

链接：https://arxiv.org/abs/2607.08234

机构：Indian Institute of Technology Indore（印度理工学院印多尔分校）

作者：Sumit Satishrao Shevtekar, Chandresh Kumar Maurya

英文摘要： Real-world time series exhibit complex dynamics characterized by multiple simultaneous temporal patterns: short-term fluctuations, periodic seasonal cycles, long-term trends, and irregular abrupt changes. However, many existing forecasting architectures rely on single-path temporal modeling--transformers capture long-range dependencies but smooth local variations, convolutions capture local patterns but have limited receptive fields, and linear models are efficient but cannot capture nonlinear dynamics. To address this, we introduce RhyMix (RHYthm MIXture), a hybrid neural architecture designed around a parallel dual-path modeling paradigm with adaptive gating mechanisms. RhyMix integrates two complementary encoding branches: (i) a Cyclic Path that incorporates explicit seasonal inductive bias through learnable cyclic embeddings, capturing predictable rhythmic patterns; and (ii) a lightweight Multi-Scale Temporal Convolutional Network with Channel Attention Path that employs multi-scale depthwise dilated convolutions to capture temporal dependencies across different receptive fields. A key innovation is the use of adaptive gating at multiple levels: a path gate dynamically combines four specialized forecasting heads (Direct, Trend-Seasonal Decomposition, Local Convolution, and Periodic Fusion) per sample and channel, while a hybrid gate adaptively balances the Cyclic and MSTCN-CA Paths based on input characteristics. This design ensures the model adapts to specific temporal patterns while maintaining linear complexity in sequence length, channels, and prediction horizon. Across extensive benchmarks on 12 real-world datasets for long-term forecasting, RhyMix achieves state-of-the-art performance on 10 of 12 datasets. The model remains lightweight (~40K params) with linear complexity and low-latency inference (<5ms),suitable for resource-constrained edge devices and real-time deployment.

AI 总结：研究语言与世界模型离散符号系统的交互问题，提出防止基于Gumbel-softmax的离散符号瓶颈失败的三个约束，经实验验证该方法能实现高基础准确率，且修复参数少、无需微调。

链接：https://arxiv.org/abs/2607.08312

作者：Jiayi Fang

英文摘要：How should language interface with a world model's discrete symbol system? The dominant paradigm -- end-to-end injection of LLM/VLM features into robot world models (RT-2, Octo, PaLM-E) -- implicitly assumes that language gradients can directly shape physical symbol representations. We ask whether this assumption is safe, find that it is not, and characterize the minimal architectural constraint that prevents the failure. Any language gradient entering a Gumbel-softmax-based discrete symbol bottleneck forces a structural trade-off: the vanilla estimator collapses to 2.2/64 symbols (4/5 seeds), while five anti-collapse strategies maintain diversity but fail to learn semantic labels (all <= 9.2% accuracy). No tested GumbelBottleneck variant achieves both objectives simultaneously. Within this family of discrete bottlenecks, the failure is structural rather than a matter of optimization. We characterize a sufficient set of three constraints that prevent the failure: (1) cut the gradient chain ( this http URL ()), preventing language signals from reaching the symbol bottleneck; (2) provide a gradient-free semantic channel -- a non-parametric Memory Table (Dict[symbol -> Counter[label]], zero parameters, zero gradients) where co-occurrence counting replaces gradient-based binding; (3) handle symbol collisions via DP-Means streaming clustering for automatic sub-cluster splitting. All three layers together achieve 97.2% grounding accuracy vs. 22.2% without Layer 3. Across two experiments spanning 74 independent runs, we demonstrate zero symbol collapse in all 32 seeds, with the blackboard achieving 79-100% semantic binding across three encoder architectures (CNN, V-JEPA 300M, CLIP ViT-L), two environments, and three texture conditions. The fix trains fewer than 2M parameters and requires no LLM fine-tuning.

AI 总结：研究探索超越反向传播的方法，提出用简单蒙特卡罗算法训练深度网络，该无梯度方法无需常见技术，灵活性高，可用于多种网络架构，为理解神经网络机制及构建深度学习系统提供新视角和途径。

链接：https://arxiv.org/abs/2607.08406

机构：Department of Physics, Xiamen University（厦门大学物理系）； Lanzhou Center for Theoretical Physics, Lanzhou University（兰州大学理论物理中心）

作者：Hong Zhao

英文摘要：Backpropagation (BP) dominates deep learning training, but its reliance on gradients brings inherent troubles -- vanishing and exploding gradients. The pursuit of gradient-free methods has long been a goal in the field of artificial intelligence. This paper shows that indeed the simplest Monte Carlo algorithm implemented on a single GPU -- randomly mutate a parameter, keep it if the loss decreases, otherwise retry -- can practically train deep networks. This gradient-free method does not even need common techniques such as batch normalization or residual connections to directly train sufficiently deep networks. More remarkably, its flexibility extends to several nontrivial scenarios: it enables pure pruning training, supports discrete weights, accommodates unconventional transfer functions such as Gaussian, and reveals the substantial redundancy of deep networks. We have demonstrated its feasibility on deep networks with more than 20 layers, single-hidden-layer wide networks with up to 16,384 hidden neurons, and even a simple Transformer architecture trained on both image classification (MNIST) and character-level language modeling (Tiny Shakespeare). This simple gradient-free method may offer a complementary perspective for understanding the self-organization and learning mechanisms of neural networks, and also provides an alternative route for building physically inspired deep learning systems.

AI 总结：针对多模态交通预测难题，提出频域多模态建模FreMo，通过模态频域滤波器细化频谱、频率引导协同积分器聚合跨模态信息，实现自适应和选择性跨模态协同，实验证明其性能优于现有基线。

链接：https://arxiv.org/abs/2607.08475

机构：Southern University of Science and Technology（南方科技大学）； The University of Tokyo（东京大学）

作者：Jiewen Deng, Hangchen Liu, Junchen Li, Boyuan Zhang, Renhe Jiang

英文摘要： Multi-modality transportation refers to urban systems composed of multiple transportation modes, such as traffic flow and public transit, whose dynamics are coupled by shared temporal patterns. Accurate multi-modality transportation forecasting remains challenging because (1) different modalities exhibit distinct spectral characteristics and (2) interact unevenly across frequencies, whereas most existing methods operate primarily in the time domain or rely on coarse feature fusion. To address these limitations, we propose a lightweight yet effective Frequency-Domain Multi-Modality modeling (FreMo) that explicitly exploits the frequency domain to enable adaptive and selective cross-modality synergy. FreMo disentangles modality-wise spectral refinement from cross-modality synergy and supports plug-and-play integration with general time series backbones. Specifically, FreMo introduces a Modality-Wise Frequency Filter (MFF) to adaptively refine spectral components within each modality, emphasizing informative frequencies while suppressing noise. FreMo further incorporates a Frequency-Guided Synergy Integrator (FSI) that selectively aggregates information across modalities based on their relative contribution at each frequency, facilitating effective cross-modality knowledge sharing while mitigating negative transfer. Extensive experiments on real-world datasets show that FreMo consistently outperforms state-of-the-art baselines, with superior performance and generalization across diverse forecasting scenarios. The code is available at this https URL.

AI 总结：研究针对主观NLP任务中注释者分歧问题，提出集成多样性优化（EDO）框架，通过统一目标联合优化集成权重等，经Gumbel-Softmax学习集成组成与大小，用有符号多样性正则化器防止集成崩溃，实验证明该方法有效提升概率校准，是模拟人类主观性的有效模型无关方法。

链接：https://arxiv.org/abs/2607.08493

机构：School of Computing and Mathematics, Manchester Metropolitan University（曼彻斯特城市大学计算与数学学院）； School of Computer Science, Hubei University（湖北大学计算机科学学院）

作者：Xia Cui, Ziyi Huang, N. R. Abeynayake

英文摘要：Subjective NLP tasks often exhibit systematic annotator disagreement, requiring models that represent uncertainty rather than collapse it. We introduce Ensemble Diversity Optimization (EDO), a prediction-space framework that jointly optimizes ensemble weights, effective cardinality, and calibration through a unified differentiable objective. EDO learns ensemble composition and size end-to-end via Gumbel-Softmax relaxation and incorporates a signed diversity regularizer, tuned on validation data, to steer optimization toward either preserving or suppressing disagreement. This regularization prevents ensemble collapse and enables controlled navigation of the utility-calibration trade-off. The framework integrates a soft F1 surrogate, class-weighted cross-entropy to address imbalance, and reliability-weighted diversity to regulate intra-ensemble variability. Experiments on four subjective text-classification benchmarks (ArMIS, ConvAbuse, HS-Brexit, MD-Agreement) show that EDO substantially improves probabilistic calibration, reducing cross-entropy (40-78% depending on baseline) and lowering Brier scores relative to Soft-CE, Soft-MD, Top-5 Voting, and WEL, while maintaining competitive F1 and better alignment with annotator distributions. These results demonstrate that jointly optimizing ensemble structure with a signed diversity regularizer provides an efficient, model-agnostic approach for modeling human subjectivity in supervised learning.

AI 总结：研究跨异构架构的学习率调度策略，通过在LEMUR数据集评估30种架构，应用25种调度器配置于九个PyTorch家族，在CIFAR-10上评估3938个模型变体，发现调度器选择依赖架构，最佳配置准确率高，为调度器选择提供参考。

链接：https://arxiv.org/abs/2607.08511

机构：Computer Vision Lab, CAIDAS & IFI, University of Würzburg（维尔茨堡大学计算机视觉实验室、CAIDAS与IFI）

作者：Hafsa Mateen, Radu Timofte, Dmitry Ignatov

英文摘要：Choosing a learning rate scheduling strategy is critical to neural network training, but manual selection is costly and rarely exhaustive. While classical AutoML approaches often treat the scheduler as a secondary hyperparameter, we systematically investigate its impact on classification accuracy across a diverse pool of architectures. We evaluated 30 representative architectures from convolutional and transformer families within the LEMUR neural network dataset. Through automated source-code injection, we applied 25 scheduler configurations across nine PyTorch families, evaluating a total of 3,938 model variants on CIFAR-10. Our best configuration achieved a top-1 accuracy of 86.45%, with 237 variants exceeding 80%. The results show that the choice of scheduler depends heavily on the architecture: CosineAnnealingWarmRestarts and CyclicLR consistently outperform basic decay strategies. The resulting accuracy landscape, contributed to the LEMUR nn-dataset, provides a practical reference for principled scheduler selection.

AI 总结：研究针对神经网络低秩正则化难题，提出简单无状态的SLORR框架，基于霍耶尔稀疏性度量和核范数有两个变体，通过GPU友好近似正则化权重矩阵，在ImageNet-1K及LLM预训练中验证其能在低开销下诱导可压缩性并保持性能。

链接：https://arxiv.org/abs/2607.08754

机构：Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）； University of Tübingen（图宾根大学）； ELLIS Institute Tübingen（图宾根ELLIS研究所）； Tübingen AI Center（图宾根人工智能中心）

作者：David González-Martínez, Shiwei Liu

英文摘要：Low-rank factorization is widely used to compress neural networks, but modern models are often not naturally amenable to aggressive factorization without significant accuracy loss. Existing training-time low-rank regularizers can improve compressibility, but they often require SVDs of large weight matrices, modify the model architecture (introducing additional trainable parameters), or rely on stateful cached quantities. To address these limitations, we introduce SLORR, a simple, stateless, and architecture-preserving framework for in-training low-rank regularization, instantiated with two main variants based on the Hoyer sparsity metric and the nuclear norm. SLORR directly regularizes the original weight matrices using GPU-friendly approximations for the forward and backward passes of the regularizers, for which we provide approximation guarantees. We first evaluate SLORR on ImageNet-1K across short-horizon continued training of ResNet-50, ViT-B/16, and ViT-L/16, and pretraining of ResNet-18, where SLORR induces compressibility while introducing less than 8% training overhead. We further evaluate SLORR-Hoyer in LLM pretraining at 135M and 560M scales: SLORR-trained compressed models preserve performance substantially better than unregularized models while adding less than 1% average training overhead.

AI 总结：介绍全睡眠模型，利用CNS/ANS划分，通过分层对比学习的三个目标进行拓扑约束表示学习，在多中心多模态PSG数据预训练后用于睡眠分期和多疾病分类，性能优于基线，凸显生理层次对睡眠表示学习的价值。

链接：https://arxiv.org/abs/2607.07720

机构：Department of Biomedical Engineering, Southern University of Science and Technology（南方科技大学生物医学工程系）； Omni-Intelligence（全知智能）； Shenzhen Loop Area Institute（深圳环区研究院）

作者： Zhoujie Hou, Song Wang, Kexin Lou, Mo Wang, Chen Wei, Quanying Liu

英文摘要：Sleep physiology arises from the coordinated dynamics of the central nervous system (CNS) and autonomic nervous system (ANS), as reflected by multimodal polysomnography signals including EEG, EOG, EMG, ECG, and respiration. However, existing sleep foundation models often fuse heterogeneous biosignals in a topology-agnostic manner, overlooking their physiological organization. We introduce Omni-Sleep, a sleep foundation model that uses the CNS/ANS partition as a physiological prior for topology-constrained representation learning. Omni-Sleep learns structured representations through three objectives: intra-system consistency, which captures shared subsystem-level factors within neural and cardio-respiratory signals; inter-system synchronization, which aligns subsystem trajectories to model brain--body dynamics; and latent-space masked temporal modeling, which captures long-horizon sleep dynamics. Pre-trained on over 100,000 hours of multi-center multimodal PSG data, Omni-Sleep is evaluated on sleep staging and multi-disease classification. Across datasets and modality-ablation settings, Omni-Sleep outperforms strong foundation-model baselines, showing improved label efficiency, cross-dataset generalization, and robustness to missing modalities. These results highlight the value of physiological hierarchy for generalizable sleep representation learning. Code is available at this https URL.

AI 总结：针对大语言模型语义嵌入特征值校准问题，提出新框架，将其与生成答案的语义嵌入视为密度矩阵预测器，通过温度缩放校准，建立相关理论并经实验验证，推动语义嵌入不确定性量化发展。

链接：https://arxiv.org/abs/2607.08377

机构：KU Leuven（鲁汶大学）； German Cancer Research Center (DKFZ)（德国癌症研究中心）； German Cancer Consortium (DKTK)（德国癌症联盟）； Goethe University Frankfurt（法兰克福歌德大学）； PSL Research University / Inria（巴黎文理研究大学/法国国家信息与自动化研究所）； Frankfurt Cancer Institute（法兰克福癌症研究所）

作者：Sebastian G. Gruber, Nassim Walha, Francis Bach, Florian Buettner

英文摘要：Uncertainty quantification is central to the reliable deployment of large language models (LLMs), and eigenvalues of semantic embeddings have recently emerged as a key tool in state-of-the-art methods. However, conventional calibration results developed for classification probabilities cannot be directly transferred to eigenvalues. We address this gap by proposing a novel framework for calibrating the eigenvalues of semantic embeddings. We interpret LLMs combined with semantic embeddings of their generated answers as density matrix predictors, and we propose a novel approach to calibrate density matrix predictors by applying temperature scaling to their eigenvalues. We establish entropy-risk equivalence under calibration, derive a central calibration inequality specific to eigenvalues, and prove that temperature-scaled eigenvalues optimize calibration when minimizing proper score risks. Experiments on a variety of real-world settings show that current LLMs are systematically overconfident, and validate our theoretical findings. Together, these results advance the foundations and practice of uncertainty quantification for semantic embeddings.

AI 总结：研究旨在整合异构材料数据以全面表征晶体材料。提出MatBind对比学习框架，以晶体结构为锚点将四种模态对齐到统一空间，实现零样本跨模态检索，且学习空间能依物理属性组织材料，组合模态可提升检索性能。

链接：https://arxiv.org/abs/2607.08470

作者：Le Yang (1), Anoop K. Chandran (2), Jona Östreicher (3), Evgenii Sovetkin (2), Adrian Mirza (4 and 6), Sebastien Bompas (1), Bashir Kazimi (1), Pascal Friederich (3), Stefan Kesselheim (2 and 7), Kevin Maik Jablonka (6, 8 and 9), Stefan Sandfeld (1 and 5) ((1) Institute for Advanced Simulations (IAS-9), Forschungszentrum Jülich,(2) Jülich Supercomputing Centre, Forschungszentrum Jülich,(3) Institute of Nanotechnology, Karlsruhe Institute of Technology,(4) Helmholtz-Zentrum Berlin für Materialien und Energie,(5) Faculty 5 - Georesources and Materials Engineering, RWTH Aachen University,(6) Helmholtz Institute for Polymers in Energy Applications Jena, (7) 1. Physikalisches Institut, University of Cologne, (8) Laboratory of Organic and Macromolecular Chemistry, Friedrich Schiller University Jena,(9) Center for Energy and Environmental Chemistry Jena, Friedrich Schiller University Jena)

英文摘要：Fully characterizing a crystalline material requires integrating heterogeneous data sources -- atomic structures, diffraction patterns, electronic density of states, and natural language -- each of which captures a different facet of the same physical object. In practice, however, these modalities are stored and analyzed in isolation, making it difficult to relate or query materials across representational boundaries. We present MatBind, a contrastive learning framework that aligns four materials modalities -- crystal structure, powder X-ray diffraction (pXRD) simulated from structures, density of states (DOS), and text -- into a unified embedding space using crystal structure as the central physical anchor. The framework induces alignment between modalities never explicitly paired during training, enabling emergent zero-shot cross-modal retrieval as a direct consequence of the shared representation. The learned embedding space organizes materials according to physically meaningful properties without explicit supervision, and retrieval performance improves systematically when modalities are combined at query time. These results demonstrate that treating heterogeneous materials data as complementary projections of a single physical reality, rather than as isolated data sources, is not a practical choice but is consistent with the underlying physics.

AI 总结：研究在强化学习智能体中建模心理障碍，将其重新表述为对认知评估信号的剂量可控操纵，通过多个旋钮表示多种障碍，呈现分级剂量反应，还发现障碍自组织成二维空间、旋钮对不同障碍有不同影响及旋钮相互作用可预测共病等，且框架具有通用性。

链接：https://arxiv.org/abs/2607.07753

作者：Hari Prasad

英文摘要： Modelling psychological disorders in artificial agents offers both a testbed for computational psychiatry and a lens on the failure modes of affective control. Prior work induces one or two disorders in a reinforcement learning (RL) agent by hand-tuned reward shaping, labels the behaviour post hoc, and reports single runs. We recast disorder modelling as dose-controllable manipulation of cognitive appraisal signals in an appraisal-guided PPO agent, expressing seven disorders (anxiety, mania, obsessive-compulsive checking, depression, impulsivity, addiction, and post-traumatic stress) each as a single knob grounded in a computational psychiatry account, with each symptom measured by a preregistered assay mapped to a recognised paradigm. Across more than a thousand runs (10 seeds, four controls, 95% confidence intervals) every disorder shows a graded, monotone dose-response that no control reproduces. Beyond these induced effects, three findings emerge that were not written into the reward: the disorders self-organise into a two-dimensional affective space in which mania mirrors anxiety; removing a knob remits reward distortion disorders (mania, checking, addiction) but not avoidance disorders (anxiety, PTSD), which instead recover under a graded exposure curriculum; and two simultaneous knobs interact nonadditively, yielding testable comorbidity predictions. Appraisal weights thus parameterise a controllable space of affective phenotypes in which the same knobs that induce a disorder can model its treatment. We also show that three disorder knobs (depression, addiction, anxiety) transfer to a three-dimensional pixel environment (MiniWorld) with a standard convolutional agent and no appraisal critic, with cross-assay dissociation confirmed across both domains, indicating the framework is not specific to grid worlds or to PPO's appraisal critic.

AI 总结：本文聚焦强化学习研究进展关键要素，分析典型评估与设计范式，介绍缩放定律理论基础，通过大规模实验表明相关研究结论有误，为深度强化学习的缩放、能力和复杂性提供核心分析。

链接：https://arxiv.org/abs/2607.07769

作者：Ezgi Korkmaz

英文摘要：Starting from the utilization of deep neural networks to approximate the state-action value function that led to winning one of the most challenging games, to algorithmic advancements that allowed solving problems without even explicitly stating the rules of the challenge at hand, reinforcement learning research has been the center of remarkable scientific progress for the past decade. In this paper, we focus on the key ingredients of this research progress and we analyze the canonical evaluation and design paradigms in reinforcement learning. We introduce the theoretical foundations of scaling laws in reinforcement learning and show that the asymptotic performance of reinforcement learning algorithms does not have a monotone relationship between performance rankings and data-regimes. We conduct large-scale experiments and our results demonstrate that a line of reinforcement learning research under the canonical design and evaluation paradigms resulted in incorrect conclusions. Our analysis and results provide a core analysis on scaling, capacity and complexity of deep reinforcement learning.

AI 总结：针对分层隐式Q学习在离线目标条件强化学习中存在的问题，提出NFTR方法，用条件归一化流取代高斯策略，结合基于架构三角不等式的三角松弛分数及RWDR目标，可避免高斯坍塌且在随机动力学下保持稳定。

链接：https://arxiv.org/abs/2607.07855

机构：Huazhong University of Science and Technology（华中科技大学）； Xi’an Jiaotong University（西安交通大学）； University of Electronic Science and Technology of China（电子科技大学）

作者：Erdemt Bao, Xing Lei, Jun Chen

英文摘要：Hierarchical Implicit Q-Learning (HIQL), an offline goal-conditioned RL method, selects subgoals by value-function advantages alone. This rule has two coupled failure modes. Optimistic bias treats lucky stochastic outcomes as skillful choices, and mode collapse reduces a multi-modal subgoal distribution to a single Gaussian mean that often falls in unreachable regions. We propose NFTR (Normalizing Flows subgoal policies with Triangle-slack Reweighting). A conditional Normalizing Flow replaces the Gaussian policy, and a closed-form mode-averaging result identifies NFs as the minimal generative class for AWR-based subgoal selection. A triangle slack score, built on the architectural triangle inequality without relying on distance accuracy, multiplicatively corrects the AWR weight to downweight subgoals whose detour cost exceeds average reachability. Triangle-slack vanishes on geodesics in deterministic MDPs and remains a conservative upper bound on composability violation under stochastic dynamics. The RWDR objective preserves AWR's population-level monotonic improvement and admits a three-term suboptimality decomposition. Together, these two ingredients yield subgoal selection that provably avoids the Gaussian collapse described above and remains stable under stochastic dynamics. GitHub page: this https URL

AI 总结：研究文本到图像合成及多模态理解中生成顺序的优化，引入经组相对策略优化训练的可学习控制模块，提升了DLMs的文本到图像对齐和多模态理解能力，在相关基准测试中取得显著改进。

链接：https://arxiv.org/abs/2607.08056

作者：Yidong Ouyang, Zhe Wang, Sourav Bhabesh, Dmitriy Bespalov

英文摘要：Diffusion Language Models (DLMs) have recently achieved substantial progress in natural language generation tasks. Recent research demonstrates that adaptive token generation ordering can significantly improve performance in mathematical reasoning and code synthesis applications. In this work, we investigate the optimization of generation order for both text-to-image synthesis and multimodal understanding. We first establish that, unlike structured problems in language generation such as Sudoku puzzles, model logits alone are insufficient for determining optimal generation sequences in text-to-image generation and multimodal understanding. To address this challenge, we introduce a learnable control module trained via Group Relative Policy Optimization (GRPO) to determine the generation order. Our results demonstrate that learning this control block substantially improves both text-to-image alignment and multimodal understanding in DLMs. In particular, it enhances the model's ability to capture fine-grained spatial relationships in generated images while also strengthening performance on multimodal reasoning and comprehension tasks. We evaluate our framework on GenEval, an object-focused benchmark for text-to-image alignment, where it achieves 4.08% relative improvements. In addition, experiments on VLMEvalKit confirm 4.85% relative improvements in multimodal understanding, highlighting the broad effectiveness of our approach.

AI 总结：研究强化学习中开放式多智能体自动课程设计难题，提出通过策略视觉检查（VIP）利用视频语言模型处理视频并提供课程建议，经星际争霸多智能体挑战赛实证，显示该方法能生成更有效课程。

链接：https://arxiv.org/abs/2607.08193

机构：Sapienza University of Rome（罗马第一大学）； Sony AI（索尼人工智能）

作者：Lorenzo Pantè, Andrea Fanti, Roberto Capobianco

英文摘要：Open-ended curricula in Reinforcement Learning (RL) aim to train generally-capable agents by identifying tasks that facilitate learning increasingly complex skills. A major challenge when designing such curricula is assessing task difficulty relative to the agent's current learning progress. While previous work has explored using scalar task scores or textual summaries of the agent's behavior, here we study a different approach: directly inspecting policy behavior via recorded episode videos. We introduce a simple yet effective instantiation of this approach which leverages a Video Language Model (VLM) to both process these videos and provide curriculum recommendations, which we call Visual Inspection of Policies (VIP). Since videos can naturally contain any number of controllable agents, we empirically study VIP on the StarCraft Multi-Agent Challenge (SMAC). We show that even with a lightweight and openly accessible VLM (VideoLLaMa2-7B), VIP can use policy videos to generate more effective curricula than both its text-only ablation and methods that rely on scalar task scores.

AI 总结：研究决斗Q学习，通过对中心化表格分解的直接解释及为非正则化、非投影的恒定步长递归建立收敛保证加强分析，推导确定性决斗Q学习的精确表示和采样随机版本的有限时间期望误差界，阐明价值和优势更新作用。

链接：https://arxiv.org/abs/2607.08340

机构：Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）

作者：Donghwan Lee

英文摘要：Q-learning is a fundamental algorithm in reinforcement learning (RL) for solving discounted Markov decision processes (MDPs) when the transition kernel is unknown. The deep Q-network (DQN) extends Q-learning by using a deep neural network for Q-function approximation, which makes Q-learning applicable to more practical high-dimensional problems. Dueling Q-learning decomposes the Q-function into a value function and an advantage function and learns the two components jointly, which can improve learning efficiency. However, the theoretical understanding of dueling Q-learning is still limited. Recent work has initiated an analysis of tabular dueling Q-learning, but existing guarantees focus on a regularized formulation and leave the pure tabular update less completely understood. This paper strengthens that line of analysis by adding a direct interpretation of the centered tabular decomposition and by establishing convergence guarantees for the unregularized, unprojected constant step-size recursion. In particular, we derive an exact switching linear system representation for deterministic dueling Q-learning and a finite-time error bound in expectation for the sampled stochastic version. The analysis clarifies how the value and advantage updates act as different gains on the action-common (value function) and action-differential (advantage function) components of the Q-function.

AI 总结：本文针对车联网运行中行为监测的挑战，提出集成强化学习、统计监测和人工反馈的在线异常检测框架，通过自注意力选择检测器、多检测器集成及人工再训练机制，在测试平台上实现持续自适应且避免灾难性遗忘。

链接：https://arxiv.org/abs/2607.08373

作者：Matthias Weiß, Athreya Hosahalli Prakash, Maurice Artelt, Falk Dettinger, Nasser Jazdi, Michael Weyrich

英文摘要：Connected vehicles are autonomous cyber-physical systems whose behavior must be continuously monitored during operation to detect deviations from normal operation before they propagate into failures. Such evaluation is challenging because the systems themselves evolve: over-the-air updates, configuration changes, and shifting workloads alter the definition of normal behavior, causing static diagnostic methods to degrade silently over time. Existing approaches typically address either automated model adaptation or operator integration in isolation, rather than as a single coordinated supervisory loop. This paper presents an online anomaly detection framework for autonomous CPS that integrates three coordinated mechanisms. A factorized deep Q-network with self-attention selects the most suitable detector from a candidate pool for each monitored service, exploiting inter-service dependencies in the microservice topology. An ensemble of three statistical drift detectors monitors the input distribution and raises an alarm only when all three concur, prioritizing precision over recall. A human-in-the-loop retraining mechanism, built around a pending transition buffer and a 60/40 prioritized replay strategy, allows the operator to incorporate expert knowledge while preserving the system's learned response to prior data distributions. The framework is evaluated on a connected-vehicle testbed running an automated valet parking application across seven backend microservices. The attention-augmented agent achieves an F1 score of 0.69, compared to at most 0.11 for any single detector applied uniformly. Following a real software update that induces measurable concept drift, F1 drops to 0.52; after operator-triggered retraining, performance recovers to 0.65 on the new distribution while remaining at 0.69 on the prior one, demonstrating sustained adaptation without catastrophic forgetting.

AI 总结：研究自主智能体在多环境下奖励学习问题，提出分层机器教学算法，先选信息丰富环境，再查询低成本反馈，实验表明该方法比统一教学基线有更低遗憾值和更强泛化能力，凸显多环境多模态教学对稳健奖励学习的重要性。

链接：https://arxiv.org/abs/2607.08647

作者：Ali Larian, Qian Lin, Chang Zong Wu, Daniel S. Brown

英文摘要： As autonomous agents are increasingly deployed across diverse operational contexts, aligning their behavior with human intent demands reward functions that remain robust to such changes rather than overfitting to any single environment. Inverse reinforcement learning (IRL) provides a principled way to infer such objectives from human feedback. However, existing analyses of optimal teaching approaches for IRL focus on single-environment, demonstration-only settings, leaving underexplored how heterogeneous feedback modalities and environment dynamics jointly constrain reward functions that generalize across multiple environments. Because demonstrations in one MDP entangle reward information with that environments specific structure, the resulting rewards frequently fail to generalize when the agent is deployed in a new setting. We first analyze how different feedback modalities constrain rewards, showing that, in the unlimited-data regime, comparisons impose strictly stronger global constraints than other modalities. Beyond this theoretical analysis, we introduce a hierarchical machine teaching algorithm for reward learning that operates across multiple MDPs. The algorithm first greedily selects informative environments that expose complementary reward constraints, then strategically queries low-cost feedback within those environments. Empirically, our method achieves substantially lower regret and stronger generalization to held-out environments than uniform teaching baselines under identical feedback budgets, demonstrating the importance of multi-environment, multi-modal teaching for learning dynamics-robust reward functions.

AI 总结：研究比特币闪电网络中给定预算下节点如何打开通道最大化路由容量的问题，采用图强化学习方法，结合消息传递策略网络、PPO和动作掩码，在真实网络快照实验中表现优于启发式基线，还用于生产中的对等推荐。

链接：https://arxiv.org/abs/2607.08703

作者：Harrison Rush, Vincent Davis, Simone Antonelli, Vikash Singh, Jesse Shrader, Emanuele Rossi

英文摘要：We address liquidity placement in the Bitcoin Lightning Network (LN): given a fixed budget, which channels should a node open to maximize its routing capacity? We cast this as a budget-constrained combinatorial optimization problem on graphs, selecting $k$ edge additions that maximize $s$--$t$ max-flow, a theory-grounded measure of routing capacity, and solve it with graph reinforcement learning. Our lightweight agent combines a message-passing policy network with proximal policy optimization (PPO) and action masking, and is trained under a hub-exclusion curriculum: the network's top hubs are removed from training subgraphs, forcing the policy to learn capacity-aware placement rather than hub attachment. In extensive experiments on real Lightning Network snapshots, our method consistently outperforms strong heuristic baselines on the max-flow objective across multiple seeds and unseen graphs. The agent has been deployed in production for peer recommendations, executing 4640 channel-open decisions that cumulatively allocate 267.3 BTC over $16 million across 30 managed nodes.

AI 总结：研究将语言模型的推理能力应用于连续控制策略的问题，提出潜在记忆宫殿（LMP）方法，通过自回归潜在空间组织信息进行变分推理，推导强化学习技术优化下限，该方法在模拟和现实领域表现良好，还产生高性能动作分词器，为控制的潜在推理提供新视角。

链接：https://arxiv.org/abs/2607.08724

机构：University of Washington（华盛顿大学）； Toyota Research Institute（丰田研究院）

作者：Chuning Zhu, Eva Xu, Jose Barreiros, Krishnan Srinivasan, Paarth Shah, Abhishek Gupta

英文摘要：Human decision-making is highly flexible -- some actions are taken immediately; others require longer deliberation. Language models have exhibited a similar capacity for adaptive "reasoning." However, transferring this capability to continuous control policies has been challenging, as directly reasoning in language space may lack the granularity for spatial understanding and precise motions. In this work, we show that reasoning for control policies can emerge by organizing information in an autoregressive latent space reminiscent of a memory palace, where retrieval is iterative and adaptive. Our method, Latent Memory Palace (LMP), formulates reasoning as variational inference with an autoregressive latent distribution. We derive a latent-space reinforcement learning technique to tractably optimize its variational lower bound. The resulting policy, LMP-$\pi$, achieves strong empirical performance in simulation and real-world domains while exhibiting interpretable, adaptive allocation of test-time compute. We further show that the same framework yields a variable-length action tokenizer, LMP-$\texttt{tok}$, which significantly improves the performance of downstream autoregressive policies. Together, these results present a new perspective on latent reasoning for control through the lens of variational inference.

AI 总结：研究扩散模型如何从有限训练集学习复杂分布，引入BIRD模型，通过贝叶斯后验推断实现时间反转扩散，确定记忆与泛化的信息理论相变边界，实验证实理论预测，揭示信息受限在生成式AI中规避维度灾难的作用。

链接：https://arxiv.org/abs/2607.08041

机构：Stanford University（斯坦福大学）

作者：Henry Hunt, Mason Kamb, Surya Ganguli

英文摘要：How diffusion models circumvent the curse of dimensionality to learn complex distributions over high dimensional spaces from a finite training set, instead of memorizing it, remains a fundamental mystery. To address this, we introduce analytically tractable Bayesian information restricted diffusion (BIRD) models, in which each pixel observes restricted information about noisy data. A BIRD model time-reverses diffusion by inferring which past training sample produced its current restricted observation using the Bayesian posterior. This model class generalizes existing analytical diffusion models that use spatially local information restriction. We show that spatially local BIRD models closely approximate trained diffusion models \textit{early in training}, across different architectures such as UNets and DiTs. Under minimal assumptions on the data distribution, we identify an information-theoretic phase boundary between memorization and generalization in the joint space of amount of training data, time in the reverse generative process, and amount of information restriction: a BIRD model memorizes when the mutual information between its restricted noisy observations and the training data exceeds the log number of training points, and it generalizes otherwise. Experiments across a range of datasets confirm our theoretically predicted location for the transition. We find that generation proceeds near the edge of memorization: both spatially local BIRD models and early-training diffusion models track the memorization-generalization phase boundary by increasingly restricting information over time. Overall, our results reveal a fundamental role for information restriction in generative AI to circumvent the curse of dimensionality.

AI 总结：该研究针对胎儿-母亲心血管系统关系建模问题提出跨模态生成框架，结合扩张卷积、跨模态注意力及自注意力，经训练可从双导联心电图合成多普勒包络，降低误差，量化母婴耦合贡献，推进了相关计算建模及胎儿评估。

链接：https://arxiv.org/abs/2607.08073

作者：Tongli Su, Alireza Rafiei, Marly van Assen, Reza Sameni, Gari D. Clifford, Faezeh Marzbanrad, Nasim Katebi

英文摘要：Fetal electrocardiogram (fECG) and Doppler ultrasound provide complementary views of fetal cardiovascular function: fECG captures electrical activity while Doppler reflects mechanical hemodynamics shaped by factors such as placental resistance and vascular compliance. Understanding the recoverable and unrecoverable Doppler components through reconstruction from fECG offers insight into the relative contributions of electrical versus mechanical factors in fetal circulation, thereby informing clinical decisions. In addition, clinical evidence of maternal-fetal cardiac coupling suggests that maternal cardiovascular dynamics may also inform fetal hemodynamics. To computationally model these relationships, we propose a cross-modal generative framework combining dilated convolutions with cross-modal attention to selectively incorporate maternal ECG and self-attention to capture long-range temporal dependencies. Trained on 885 synchronized fetal/maternal ECG and Doppler envelope segments from 39 pregnancies, our model synthesizes Doppler envelopes with power spectral density mean squared error (PSD MSE) of 49.9 +/- 15.8 dB^2 (51% lower than two-channel baseline) and heart-rate error of 4.71 +/- 0.77 bpm (1.5% better than baseline; negligible relative to the 110-160 bpm physiological range). Cross-modal attention yields a 39% PSD MSE reduction over naive dual-channel concatenation, quantifying the contribution of maternal-fetal coupling. Our proposed framework advances computational modeling of the maternal-fetal cardiovascular system by enabling the synthesis of Doppler envelopes from dual-lead ECG. By analysis of both recoverable and residual Doppler components, this approach enables quantification of the purely mechanical contributions to Doppler waveforms -- those not recoverable from electrical recordings -- ultimately facilitating a more comprehensive fetal assessment.

AI 总结：针对缺乏配对数据及目标变异异质时现有方法的不足，提出CASL-VAE模型，它能从不配对数据学习结构化潜在因子，经变分推理实现相关优化，在半合成神经影像数据验证中表现良好，可用于半监督聚类、配对样本生成及揭示疾病异质性。

链接：https://arxiv.org/abs/2607.08254

作者：Sai Spandana Chintapalli, Pratik Chaudhari, Christos Davatzikos

英文摘要：Quantifying variability in a target population relative to a reference population is central to many scientific and clinical problems (e.g., diseased vs. healthy). Yet, without paired data and in the presence of heterogeneous target variation, existing methods struggle to separate multiple modes of target-specific variation. We propose \textit{CASL-VAE}, a deep contrastive latent variable model that learns structured latent generative factors from unpaired data. CASL-VAE factorizes variation into continuous common latent factors shared across populations and hierarchical salient latent factors that model target-specific heterogeneity as discrete subtypes and continuous within-subtype variation. Using variational inference, we show how approximate joint likelihood optimization over reference and target domains can be performed using unpaired data, providing a principled basis for paired-sample generation and cross-domain analysis. We validate CASL-VAE on semi-synthetic neuroimaging data, demonstrating improved subtype recovery and paired-sample generation compared to baseline clustering and generative models. We also validate its ability to reveal biologically plausible heterogeneity in Alzheimer's disease.

AI 总结：研究文本到图像模型中扩散去学习的不稳定问题，提出AutoAnchor两阶段框架，利用交叉注意力一致性损失自动合成流形近端锚点，有效实现稳健无偏去学习，提升目标概念去除率和非目标效用，还可集成到现有方法中提高性能。

链接：https://arxiv.org/abs/2607.08337

机构：Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））

作者：Siyuan Wen, Jiahao Zeng, Ningning Ding

英文摘要：Diffusion unlearning is essential for mitigating the generation of harmful or copyrighted content in text-to-image models. Current diffusion unlearning techniques determine the model update direction by either using alternatives of the target concept as an anchor or using empty prompts. The anchor-based method relies on manually and semantically-chosen anchors that risk biased unlearning, while the anchor-free method inherently suffers from unrobust unlearning due to unconstrained latent updates. In this work, we theoretically formalize such unstable diffusion unlearning issues under the manifold hypothesis and prove that lacking a manifold-proximal anchor inevitably induces significant normal-space drift that degrades unlearning performance. To achieve stable unlearning, we propose \mysysn, a two-stage framework that automatically synthesizes manifold-proximal anchors. However, direct geometric manifold optimization is computationally intractable. To address this challenge, \mysys introduces a novel cross-attention consistency loss which serves as a highly efficient surrogate of manifold proximity. Experimental results demonstrate that \mysys effectively achieves robust and unbiased unlearning across various state-of-the-art baselines, significantly improving targeted concept removal (by up to 31.04\% in CLIP score) and non-target utility (by up to 4.18\% in CLIP score). Moreover, \mysys can also be easily integrated into existing diffusion unlearning methods to enhance their unlearning performance (by 6.30\% for concept removal and 6.65\% for utility on average).

AI 总结：研究两层神经网络拟合含噪声标签时的Lipschitz常数，证明对任意实权重等情况，在高概率下拟合低于噪声底限的数据会使\(\mathrm{Lip}(f)\)满足特定不等式，通过函数空间覆盖及刚性引理完成证明，解决了无界权重下的问题。

链接：https://arxiv.org/abs/2607.07778

机构：Einstein Institute of Mathematics, The Hebrew University of Jerusalem（耶路撒冷希伯来大学爱因斯坦数学研究所）

作者：Yitzchak Shmalo

英文摘要： Bubeck, Li and Nagaraj conjectured that, for generic data, any two-layer neural network with $m$ neurons that fits $n$ noisy labels must have Lipschitz constant at least of order $\sqrt{n/m}$, with no restriction on the size of the weights. Bubeck and Sellke proved a universal version of this law for Lipschitz-parameterized classes, but under a polynomial bound on the parameters; at depth three that boundedness hypothesis is genuinely necessary. The two-layer unbounded-weight case requires a different argument. We prove the conjectured law, up to one logarithmic factor, for every continuous piecewise-linear activation, in particular for ReLU networks. For data drawn uniformly from $\mathbb{S}^{d-1}$, $d\ge3$, or from $N(0,I_d/d)$, labels in $[-1,1]$ with noise level $\sigma^2>0$, and any width-$m$ two-layer network with arbitrary real weights, biases and affine skip connection, fitting the data $\varepsilon$ below the noise floor forces $\mathrm{Lip}(f)\ge c\,\varepsilon\sqrt{n/(\bar m\log(C\bar m nd/\varepsilon))}$, $\bar m=(K-1)m+1$, with high probability. A realized-kink-count version holds on the same event: every realized two-layer piecewise-linear function with $k(f)\le n$ distinct kink hyperplanes obeys the bound with $\bar m$ replaced by $k(f)+1$, irrespective of how many redundant hidden units parameterize it. The proof replaces parameter-space covering, impossible for unbounded weights, by a function-space covering. The central deterministic ingredient is a rigidity lemma: on $B_2$, and on $\mathbb{S}^{d-1}$ for $d\ge3$, the coefficient of each canonical kink is controlled by the Lipschitz constant of the realized function, because kinks on distinct hyperplanes cannot cancel at generic points. Rigidity genuinely fails at $d=2$, and an explicit two-layer ReLU interpolant with $O(1)$ Lipschitz constant at width $2n$ matches the law at the overparameterized endpoint.

AI 总结：研究神经网络海森矩阵接近零的特征值的起源，通过分析网络参数化连续对称性的弱提升赝戈德斯通模式，结合ReLU微扰及特征向量频谱解析，在多种模型中展示机制，将海森矩阵主体与弱破缺对称性相联系并阐明接近零模式起源。

链接：https://arxiv.org/abs/2607.07845

作者：Marcel Kühn, Bernd Rosenow

英文摘要：The Hessian of the training loss governs the local geometry of the loss landscape, yet despite existing explanations for its largest eigenvalues, the origin of the vast multitude of vanishingly small eigenvalues remains elusive. We argue that the bulk consists of the weakly lifted pseudo-Goldstone modes of the continuous symmetries of the network parametrization. In deep linear networks these symmetries are exact: they generate flat directions and hence exact zero modes, whose eigenvectors we construct explicitly. Introducing a ReLU nonlinearity as a perturbation, we show that it breaks these symmetries weakly and explicitly. Resolving the spectrum at the level of eigenvectors, we find that the high-curvature directions are orthogonal to the symmetry subspace, while the bulk lies almost entirely within it. We demonstrate the mechanism in a two-layer ReLU student--teacher model and in a network trained on CIFAR-10. A convolutional example demonstrates that the same diagnostic extends beyond fully connected layers. Together, these results link the Hessian bulk to weakly broken symmetries and clarify the origin of near-zero modes.

AI 总结：研究辅助博弈在线变体中信息完备与不完备智能体交互优化奖励函数的问题，提出人类和助手的分散算法，实现(1 - 1/e)近似辅助遗憾率为O(T^(3/4))，证明更好因子难处理，还展示算法在伪分散设置下达O(T^(1/2))速率。

链接：https://arxiv.org/abs/2607.08012

作者：Nivasini Ananthakrishnan, Mark Bedaywi, Michael I. Jordan, Stuart Russell, Nika Haghtalab

英文摘要：This paper studies an online variant of the assistance games framework, where an informed agent and an uninformed agent repeatedly interact over $T$ timesteps to optimize a common reward function. While the informed agent (the human) observes a latent state of the world, the uninformed agent (the assistant) observes only the human's actions. We provide the first provably efficient learning algorithms for repeated assistance games. We introduce the notion of assistance regret: the gap between the cumulative utility of interactions and that of the optimal joint policies in hindsight, which map latent states to action pairs. We present decentralized algorithms for both the human and the assistant that achieve a $(1-1/e)$-approximate assistance regret rate of $\widetilde{O}(T^{3/4})$, with runtime polynomial in the size of the action and state spaces. These algorithms are general; in particular, they accommodate any no-regret algorithm for the assistant. We prove that achieving a regret approximation factor better than $(1-1/e)$ is computationally intractable. Furthermore, we demonstrate how these generic no-regret algorithms can be tailored to a pseudo-decentralized setting -- using a shared random string -- to achieve a rate of $\widetilde{O}(T^{1/2})$, optimal up to logarithmic factors.

AI 总结：研究反射布朗运动平稳分布，基于基本伴随关系开发深度学习方法学习其拉普拉斯变换，精心设计框架各部分，在已知尾概率的实例上评估，结果近乎完美，有望成为分析随机系统的通用工具。

链接：https://arxiv.org/abs/2607.08091

机构：Cornell University（康奈尔大学）

作者：Jim Dai, Zhanhao Zhang

英文摘要：The stationary distribution of reflected Brownian motion (RBM) plays an important role in the analysis of high-dimensional stochastic systems, yet closed-form solutions are known only for a few special cases. Computing important performance metrics, such as tail probabilities, is even more intractable, despite their practical relevance. In this paper, we develop a deep learning approach that accurately and efficiently learns the Laplace transform of high-dimensional RBMs based on the basic adjoint relationship (BAR). Our framework combines a careful design of the loss function, training data sampling procedure, and neural network architecture. We evaluate the proposed method on RBM instances with known ground-truth tail probabilities and demonstrate near-perfect prediction in high-dimensional settings, highlighting its potential as a general tool for analyzing stochastic systems beyond analytically tractable regimes. Our code can be found at this https URL.

AI 总结：研究标签噪声下的秩估计问题，提出随机序学习（SOL）框架，通过判别损失和随机序损失两个互补目标捕捉不确定性并学习嵌入空间，实验证明该框架能在多种标签噪声下实现可靠秩估计。

链接：https://arxiv.org/abs/2607.08103

作者：Chaewon Lee, Seon-Ho Lee, Chang-Su Kim

英文摘要： Rank estimation under label noise poses a fundamental challenge, as ordinal annotations often exhibit structured uncertainty rather than simple label corruption. In this paper, we reformulate rank estimation with noisy ordinal labels as a stochastic ordering problem, in which each instance is inherently associated with multiple plausible ranks instead of a single deterministic label. Based on this view, we propose stochastic order learning (SOL), a learning framework that captures ordinal label uncertainty and learns an embedding space through two complementary objectives: a discriminative loss that structures instance--centroid interactions and a stochastic order loss that enforces probabilistic ordering relations between instances. Extensive experiments across diverse datasets demonstrate that SOL enables reliable rank estimation under various types and levels of label noise. The source code is available at this https URL.

AI 总结：研究普通SGD尤其是带动量的在重尾噪声下的表现，通过细化收敛结果及全面分析，揭示其在重尾噪声下收敛率低于裁剪或归一化变体，展现普通方法局限性，合成函数实验支持理论发现。

链接：https://arxiv.org/abs/2607.08104

机构：Meiji University（明治大学）

作者：Ryusei Yamada, Naoki Sato, Hideaki Iiduka

英文摘要：Stochastic gradient descent (SGD) is a cornerstone of modern optimization. While its performance under heavy-tailed noise is often addressed through specialized modifications such as gradient clipping or normalization, we investigate a more fundamental question: how does vanilla SGD, particularly with momentum, perform in the presence of heavy-tailed noise? In this paper, we refine existing convergence results for vanilla SGD and, more importantly, provide the first comprehensive convergence analysis of vanilla SGD with momentum for strongly convex, convex, and nonconvex objectives, without employing any gradient control mechanisms. Our results demonstrate that the obtained convergence rates are inferior to the optimal rates achieved by clipped or normalized variants of SGD, thereby revealing inherent limitations of vanilla methods under heavy-tailed noise. The theoretical findings are supported by experiments on synthetic functions.

AI 总结：该研究提出对比顺序学习框架ConOrd，整合对比学习与顺序学习优势，通过引入基于秩差的对比顺序损失解决相关局限，在面部年龄估计等序数回归任务实验中达最优性能且泛化性好。

链接：https://arxiv.org/abs/2607.08109

作者：Chaewon Lee, BeomJun Shim, Kwang Pyo Choi, Chang-Su Kim

英文摘要：We propose contrastive order learning (ConOrd), a contrastive learning framework for ordinal regression that integrates the strengths of contrastive learning and order learning. While contrastive learning effectively leverages all samples in a batch, it typically ignores the inherent ordering among rank labels. Conversely, order learning explicitly models label ordinality but often relies on local, margin-based comparisons, limiting its ability to capture global ordinal structure. ConOrd addresses these limitations by introducing a contrastive order loss with soft affinity and disparity weights based on rank differences, enabling fine-grained modeling of ordinal relationships across all sample pairs within a batch. Extensive experiments on a range of ordinal regression tasks, including facial age estimation, blind image quality assessment, and blind video quality assessment, demonstrate that ConOrd consistently achieves state-of-the-art performance and generalizes well across diverse ordinal regression scenarios. The source code is available at this https URL.

AI 总结：研究k均值++算法重启次数问题，提出结合古德-图灵估计等的GTRC准则，该准则能依数据集难度调整重启次数，使聚类质量具竞争力，为固定重启次数提供可解释且有原则的替代方案。

链接：https://arxiv.org/abs/2607.08243

作者：Renato Cordeiro de Amorim

英文摘要：The k-means++ algorithm is commonly restarted multiple times to avoid poor local optima, yet the number of restarts is almost always chosen arbitrarily and applied uniformly regardless of data set difficulty. This undermines any comparison relying on such a choice and wastes computation on easy data sets while potentially under-serving hard ones. We introduce GTRC, a restart criterion combining a Good-Turing estimate, a proven unconditional bound, and a confidence-based bound on the probability that a further restart would improve on the current result, stopping once this probability falls below a user-specified tolerance $\varepsilon$. Across 36 data sets, GTRC reached clustering quality competitive with well-chosen fixed restart counts, while the number of restarts used varied considerably and appropriately with data set difficulty, governed by an interpretable, data-dependent signal rather than a fixed rule. GTRC offers a principled and reportable alternative to fixing the number of $k$-means++ restarts in advance. Software: this https URL.

AI 总结：研究在允许有效局部采样器的图形模型下学习 $\mathsf{AC}^0$ 的问题，提出通过模拟和截断经典格劳伯动力学建立新的低阶近似的方法，得到拟多项式时间学习者，还能应用于两自旋系统。

链接：https://arxiv.org/abs/2607.08303

作者：Weiming Feng, Xiongxin Yang, Yixiao Yu, Yiyao Zhang

英文摘要： The problem of learning constant-depth circuits holds profound implications for computational learning theory. In a seminal result, by introducing the low-degree algorithm, Linial, Mansour, and Nisan (J. ACM 1993) presented a quasipolynomial-time learner for $\mathsf{AC}^0$ under the uniform distribution. However, obtaining comparable learning guarantees for broader classes of correlated distributions has remained a longstanding challenge. Recently, Chandrasekaran, Gaitonde, Moitra, and Vasilyan (arXiv 2026) extended these guarantees to Gibbs distributions on bounded-degree graphical models with both strong spatial mixing and polynomial growth. In this paper, we give a quasipolynomial-time learner for $\mathsf{AC}^0$ under graphical models that admit efficient local samplers, circumventing the polynomial-growth requirement in prior work. The key ingredient is a new low-degree approximation for Gibbs distributions, established by simulating and suitably truncating the classical Glauber dynamics. As applications, this framework yields learners for two-spin systems, including the hard-core model and Ising model, on arbitrary bounded-degree graphs, in regimes approaching their respective sampling thresholds.

AI 总结：本文将梯度下降理论扩展到向量值输出的超参数化最小二乘及平坦极小值流形邻域，推广了范式和收敛定理，克服技术挑战，还表明框架适用于深度矩阵分解，得出新结构结果。

链接：https://arxiv.org/abs/2607.08380

作者：Lachlan Ewen MacDonald, René Vidal

英文摘要：An important quantity in the theory of gradient descent (GD) is the \emph{sharpness}, defined as the largest eigenvalue of the objective Hessian. Classical analyses typically require the step size to be uniformly smaller than twice the reciprocal of the sharpness, but this condition is frequently violated in the training of deep neural networks. Recent work bridges this gap in the setting of overparametrised least-squares with a \emph{single scalar output}, providing a normal form for large-step GD in a neighbourhood of an \emph{isolated} flat minimum and establishing three corresponding convergence results. In this paper, we extend this theory in two directions: (1) to overparametrised least-squares with \emph{vector-valued outputs} (including regression with arbitrarily many observations), and (2) to a neighbourhood of a \emph{manifold} of flat minima (which we show is essential for applications such as matrix factorisation). We generalise both the normal form and all three convergence theorems of \cite{macdonaldeos} to this broader setting, overcoming several technical challenges, including the solution of a singular partial differential equation via a novel method that may be of independent interest. We further show that our framework applies to deep matrix factorisation under mild assumptions, yielding several new structural results. In particular, we prove that the set of flat minima forms a fibre bundle over a product of spheres, and that the sharpness is Morse-Bott along this manifold.

AI 总结：研究基于伪逆的极限学习机的谱稳定性，通过分析最小奇异值和条件数来探讨其稳定性，比较基于奇异值分解和迭代超幂方法，实验表明病态条件下基于奇异值分解的方法更可靠，揭示ELM稳定性由隐藏层矩阵奇异值结构决定。

链接：https://arxiv.org/abs/2607.08581

作者：Bich Van Nguyen, Ngoc Anh Khong

英文摘要：Extreme Learning Machine (ELM) computes output weights analytically using the Moore-Penrose pseudoinverse. Although this leads to fast training, its numerical stability depends strongly on the conditioning of the hidden layer matrix. This paper studies pseudoinverse-based ELM from a spectral perspective. We show that the smallest singular value governs perturbation amplification in the output weights, while the condition number provides a quantitative measure of hidden-layer instability. We compare SVD-based pseudoinverse computation with iterative hyperpower methods and discuss width-dependent conditioning through a random feature interpretation. Experiments on synthetic matrices and ELM benchmarks show that SVD-based methods remain the most reliable under ill conditioning, while iterative methods are more sensitive to spectral properties. The results suggest that ELM stability is fundamentally governed by the singular value structure of the hidden layer matrix.

AI 总结：研究提出KronQ框架，通过引入梯度协方差挑战现有二阶PTQ方法假设。利用克罗内克分解海森矩阵近似，在两个层面改进量化：双向非相干处理减少权重幅度方差，导出新敏感度度量用于层间混合精度分配，在特定量化任务中表现优于GPTQ等。

链接：https://arxiv.org/abs/2607.07964

作者：Donghyun Lee, Yuhang Li, Ruokai Yin, Priyadarshini Panda

英文摘要：Post-training quantization (PTQ) is a widely adopted technique for compressing large language models (LLMs) without retraining. Existing second-order PTQ methods, including GPTQ, construct quantization objectives exclusively from input activation statistics, effectively assuming that all output channels contribute equally to the layer-wise reconstruction objective. We propose KronQ, a PTQ framework that challenges this assumption by introducing the gradient covariance into the quantization pipeline. Under the Kronecker-factored Hessian approximation, the quantization loss depends jointly on both the activation and gradient covariances, and KronQ exploits this at two complementary levels. (1) KronQ introduces bidirectional incoherence processing, extending the existing input-side random rotation to the output dimension using the gradient covariance, reducing weight magnitude variance across both input and output dimensions. (2) KronQ derives a new sensitivity metric for inter-layer mixed-precision allocation, driven by the gradient and activation Hessian traces. Notably, in the case of 2-bit weight-only quantization on LLaMA-3-70B, while GPTQ and GPTAQ diverge or produce degenerate quantizations (>2000 perplexity on WikiText-2), KronQ achieves 7.93 perplexity.

AI 总结：本文针对小型视觉语言模型量化，提出系统评估框架，在特定硬件上验证五个假设，分离相关组件得出结论，包括量化敏感性受结构范式影响、SigLIP编码器延迟情况、大语言模型INT4量化利弊、复合量化误差特点及不同平台每焦耳智能差异。

链接：https://arxiv.org/abs/2607.08029

作者：Hyeju Shin, Chorwon Kim, Ryangsoo Kim, Hark Yoo, Jaein Kim

英文摘要： The emergence of vision language models with fewer than 3 billion parameters has accelerated the implementation of on-device multimodal intelligence. However, a detailed understanding of component-wise quantization remains a bottleneck for optimal deployment. This paper presents a systematic evaluation framework for empirically validating five hypotheses across six quantization configurations on the Jetson Orin NX and AGX. By separating the vision encoder, projector, and large language model backbone yields the following results: (1) Quantization sensitivity is governed by the structural paradigm (MoE vs. dense) rather than scale alone, with MoE backbones mitigating INT4 noise where dense backbones degrade; (2) SigLIP encoders incur disproportionate INT8 latency on Jetson Ampere--a deployment-specific encoder-kernel-hardware interaction, not a SigLIP flaw; (3) Although INT4 quantization of LLMs greatly reduces VRAM consumption, it also causes slower token generation due to dequantization overhead; (4) Composite quantization errors are largely additive, except along the modality-alignment path, which is architecture-dependent; (5) The intelligence-per-joule profile varies significantly across platforms owing to memory bandwidth constraints.

AI 总结：该综述聚焦大语言模型服务中系统感知的键值缓存优化，从执行与调度、放置与迁移、表示与保留三个维度回顾相关工作，分析跨行为协同设计及行为与目标联系，为理解和创新键值缓存设计提供基础。

链接：https://arxiv.org/abs/2607.08057

机构：School of Computing and Information Systems, The University of Melbourne（墨尔本大学计算与信息系统学院）； School of Computer Science and Technology, Huazhong University of Science and Technology（华中科技大学计算机科学与技术学院）

作者：Jiantong Jiang, Peiyu Yang, Rui Zhang, Feng Liu

英文摘要：Despite the rapid advancements of large language models (LLMs), LLM serving systems remain memory-intensive and costly. The key-value (KV) cache, which stores KV tensors during autoregressive decoding, is crucial for enabling low-latency, high-throughput LLM inference serving. In this survey, we focus on system-aware KV infrastructure for serving LLMs (abbreviated as sKis). We revisit recent work from a system behavior perspective, organizing existing efforts into three dimensions: execution and scheduling (temporal), placement and migration (spatial), and representation and retention (structural). Furthermore, we analyze cross-behavior co-design affinity and behavior-objective links, highlighting future opportunities. Our work systematizes a rapidly evolving area, providing a foundation for understanding and innovating KV cache designs in modern LLM serving infrastructure.

AI 总结：研究大语言模型部署时的压缩问题，提出BiSCo-LLM框架，通过将局部权重块映射到超球面二值化、编码重建误差、进行类别恢复蒸馏及设置保护通道路径等方法，实现极低比特权重压缩并减少存储开销。

链接：https://arxiv.org/abs/2607.08643

作者：Yuantian Shao, Peisong Wang, Zhilei Liu, Chuangyi Li, Yuanteng Chen, Pengcheng Xie, Yiwu Yao, Zhihui Wei, Jian Cheng

英文摘要：Large language models (LLMs) are increasingly constrained by memory capacity, weight bandwidth, and checkpoint storage during deployment. Existing low-bit compression methods mainly follow two directions. Scalar or group-wise quantization is simple and compatible with efficient low-precision kernels, but its representation capacity becomes limited when the target budget approaches 2 bits per weight. Vector-quantized weight compression provides a richer block-level representation, but usually introduces explicit codebooks, index lookup, and additional storage accounting. This paper presents BiSCo-LLM, a codebook-free binary spherical coding framework for extreme low-bit LLM weight compression. The core pipeline is built on three components. First, local weight chunks are mapped onto a unit hypersphere and binarized into compact spherical codes, so that the main payload is a bit-packed sign stream rather than explicit VQ centroids. Second, a residual BSQ stage encodes the reconstruction error left by the base spherical codec, providing an explicit rate-distortion path without stored codebooks. Third, category-wise recovery distillation is performed after replacing each Transformer module category, reducing the mismatch between local weight reconstruction and assembled model behavior. A small 8-bit protected-channel path is used as an auxiliary stabilization mechanism for sensitive channels and is counted separately from the BSQ payload. The reported storage budget includes binary codes, neural decoders, protected-channel payloads, LoRA adapters, and metadata.

AI 总结：针对实时边缘系统中FL受设备异质性影响、推理速度慢的问题，提出Collate框架，设计动态归零恢复方法和原型校正联邦聚合方案，实验表明该框架能在延迟约束下有效提升模型准确率且训练开销小。

链接：https://arxiv.org/abs/2607.08013

机构：NTU（南洋理工大学）； HP Inc.（惠普公司）

作者：Shuo Huai, Di Liu, Hao Kong, Xiangzhong Luo, Weichen Liu, Ravi Subramaniam, Christian Makaya, Qian Lin

英文摘要：Federated Learning (FL) empowers multiple clients to collaboratively learn a model, enlarging the training data of each client for high accuracy while protecting data privacy. However, when deploying FL in real-time edge systems, the heterogeneity of devices among systems has a severe impact on the performance of the inferred model. Existing optimizations on FL focus on improving the training efficiency but fail to speed up inference, especially when there is a latency constraint. In this work, we propose Collate, a novel training framework that collaboratively learns heterogeneous models to meet the latency constraints of multiple edge systems simultaneously. We design a dynamic zeroizing-recovering method to adjust each local model architecture for high accuracy under its latency constraint. A proto-corrected federated aggregation scheme is also introduced to aggregate all heterogeneous local models, satisfying the latency constraint of different systems with only one training process and maintaining high accuracy. Extensive experiments indicate that, compared to state-of-the-art methods and under a latency constraint, our extended models can improve the accuracy by 1.96% on average, and our shrunk models can also obtain a 3.09% accuracy improvement on average, with almost no extra training overhead. The related codes and data will be available at this https URL

AI 总结：研究针对因果推断的差分隐私合成数据，提出因果工作负载，围绕双稳健因果估计器的正交矩设计DP查询集，经最大熵校准重建合成数据，还引入相关工具，能在不额外隐私支出下支持多种分析，且揭示了分布保真度与因果推断的权衡。

链接：https://arxiv.org/abs/2607.08122

作者：Amir Asiaee, Kaveh Aryan

英文摘要：Workload-based differentially private (DP) synthetic data methods privately measure aggregate queries and post-process the noisy answers into synthetic records. Generic workloads can achieve strong distributional fidelity, but causal estimands such as the average treatment effect (ATE) depend on treatment-arm balance and outcome moments that generic marginals need not preserve. We propose causal workloads: DP query sets designed around the orthogonal moments used by doubly robust causal estimators. The released workload can be used directly by stable moment-map estimators or reconstructed by maximum-entropy calibration into reusable synthetic data; our theory decomposes ATE error into sampling, privacy, workload-approximation, Monte Carlo, and calibration terms. We also introduce Causal-AIM, an adaptive workload selector, and a noise-aware multiple-imputation (NA+MI) procedure for confidence intervals from DP synthetic data. Because the workload is released once, the same DP synthetic table can support ATE, ATT, and subgroup analyses without additional privacy spending. Empirically, causal workloads are most useful at strict privacy budgets and for calibrated uncertainty, while generic workloads often retain an advantage for point RMSE as privacy relaxes. The broader lesson is a tradeoff: distributional fidelity can help point accuracy, but valid causal inference requires preserving causal moments and propagating DP noise rather than treating synthetic rows as real.

AI 总结：研究针对跨机构心血管疾病风险预测受隐私限制问题，提出联邦深度学习方法，整合不同特征队列，经实验，该方法训练的模型比本地训练模型预测性能更高，提升了鹿特丹研究和生命线研究的C统计量，保护了患者隐私。

链接：https://arxiv.org/abs/2607.08595

作者：Hyunho Mo, Djura Smits, Mahlet A. Birhanu, Maarten J.G. Leening, Daniel Bos, Pim van der Harst, Esther E. Bron

英文摘要：Cardiovascular disease risk prediction models often rely on data from a single institution or centrally pooled datasets. Extending these models across institutions could be limited by privacy regulations and constraints on sharing patient-level data. Federated learning enables collaborative model development without transferring sensitive patient data, but its application in healthcare remains challenging because datasets often differ in size, population characteristics, and outcome definitions. In this study, we present a federated deep learning approach for privacy-preserving cardiovascular disease risk prediction that integrates two population-based cohorts with different characteristics: Lifelines, including 148,230 participants meeting the study inclusion criteria with self-reported outcomes, and the Rotterdam Study, including a smaller cohort of 10,155 participants with digitally linked clinical outcomes. Model performance was primarily evaluated on the Rotterdam Study because of its complete follow-up. Deep survival models trained using federated learning achieved higher predictive performance than models trained locally without federation. For the Rotterdam Study, the C-statistic increased from 0.728 (95% CI: 0.717-0.739) to 0.739 (95% CI: 0.728-0.749). For Lifelines, the C-statistic increased from 0.783 (95% CI: 0.775-0.791) to 0.787 (95% CI: 0.780-0.792). These findings suggest that federated deep learning across heterogeneous cohorts can improve cardiovascular disease risk prediction while preserving the privacy of individual-level patient data.

AI 总结：研究如何实现安全的去中心化联邦学习，提出gspDAG-FL框架，通过八卦历史达成共识，结合多种验证方法提高弹性，经实验验证其在减少协调瓶颈、提高吞吐量及检测无效起源等方面效果良好，学习质量接近基于验证的账本FL。

链接：https://arxiv.org/abs/2607.08651

机构：Columbia University（哥伦比亚大学）

作者：Amirhossein Taherpour, Xiaodong Wang

英文摘要：Decentralized federated learning (DFL) removes the central server by letting nodes exchange model updates through peer-to-peer gossip, but existing gossip-based methods often lack provenance finality and resilience to Byzantine or lazy participants. Ledger-assisted federated learning (FL) improves auditability, yet blockchains, shards, or settlement committees can reintroduce global coordination costs that conflict with DFL locality. This paper proposes \emph{gspDAG-FL}, a secure DFL framework that derives consensus from the same gossip history used to disseminate models. Nodes exchange model payloads only with neighbors, while full nodes collect event certificates and receiver-endorsed accepted gossip proofs, reconstruct a compact Topology directed acyclic graph (DAG), and run Hashgraph-style virtual voting followed by compact full-node certificates. Finality is over unique model-origin tuples, not identical local parameter states. To improve resilience, gspDAG-FL combines payload validation, accepted-proof validation, and private semantic audit before aggregation. We formalize the adversarial setting, prove safety and conditional liveness of the control plane, and give a convergence guarantee for certified perturbed gossip under time-varying effective mixing. Experiments on MNIST classification and Penn Treebank language modeling, using fair held-out validation/audit data and networks up to \(N=100\), show that gspDAG-FL achieves learning quality close to validation-based ledger FL while reducing coordination bottlenecks, improving throughput, and maintaining high invalid-origin detection under mixed Byzantine and lazy participation.

AI 总结：研究针对图神经网络在隐私敏感领域应用难题，提出EdgeRefine框架，利用杰卡德采样和隐私预算确定边比例并采样，在跨数据集和架构上提升隐私-效用平衡，实验表明其准确率高且抗隐私泄露能力强。

链接：https://arxiv.org/abs/2607.08659

机构：School of Cyber Engineering, Xidian University（西安电子科技大学网络与信息安全学院）； State Key Laboratory of Integrated Services Networks（综合业务网理论及关键技术国家重点实验室）

作者：Wenxiu Ding, Muzhi Liu, Zheng Yan, Mingjun Wang, Yifan Zhao, Qiao Liu

英文摘要： Graph Neural Networks (GNNs) have shown considerable success in learning from graph-structured data, but their use in privacy-sensitive areas remains difficult because graph structure can leak sensitive link information. To satisfy edge-level differential privacy, a common approach is to inject noise into all elements of the graph's adjacency matrix, thereby obfuscating the existence of any single edge. However, stronger privacy requires more noise, and excessive noise reduces utility, making the privacy-utility balance a major barrier to practical privacy-preserving graph learning. To address this issue, we propose EdgeRefine, a local differential privacy framework that improves this trade-off through adaptive edge refinement. EdgeRefine first estimates edge-existence probabilities using Jaccard similarity and ranks edges for noisy edge removal. To ensure the sparsity and reliability of the final graph, it uses the privacy budget $\epsilon$ to determine the ratio of true to false edges, samples them separately based on this probability ranking, and controls the total number of edges with a separate sampling rate $k$. Extensive experiments show that EdgeRefine achieves accuracy comparable to the noise-free baseline and substantially outperforms other privacy-preserving methods across datasets and GNN architectures. Under privacy budget $\epsilon = 2.5$, EdgeRefine improves node classification accuracy over state-of-the-art baselines by 17.8\% on ACM under GAT and 19.7\% on Cora under GCN. In graph classification, it achieves an average accuracy degradation of around 5\% compared to the noise-free baseline. Under graph reconstruction attacks, EdgeRefine maintains relative absolute error levels above 1 across all privacy budgets, averaging 1.962 on Cora and 1.472 on AMAP, indicating strong resilience against privacy leakage.

AI 总结：研究旨在开发同时满足准确性、鲁棒性和校准的神经网络。核心方法是引入LiST训练范式，通过迭代调整全局Lipschitz常数达到理想操作点，并利用训练损失中的余量参数构建校准前沿。主要贡献是在多数据集上验证该方法，展现出良好性能。

链接：https://arxiv.org/abs/2607.07745

作者：Arthur Chiron (IRIT, EPE UT), Franck Mamalet, Thomas Massena (IRIT, DTIPG - SNCF, UT3), Thomas Deltort (IRIT), Mathieu Serrurier (IRIT, UT2J)

英文摘要：While accuracy, robustness, and calibration are all essential for reliable neural networks, they are often studied separately; developing models that satisfy all three simultaneously remains a central challenge. Lipschitz-constrained models guarantee robustness by design, yet the manual selection of the Lipschitz constraint L governs the resulting accuracy-robustness trade-off, and their calibration properties remain largely underexplored. In this work, we highlight a theoretical and empirical link between the enforced Lipschitz constraint and Temperature Scaling, a state-of-the-art calibration method. Specifically, we find that for a given training scheme, there exists a non-trivial value L* that yields an out-of-the-box calibrated network, and that calibration acts as a principled criterion to select a well-defined operating point on the accuracy-robustness Pareto front. Leveraging these insights, we introduce Lipschitz Scaling Training (LiST), a novel training paradigm that iteratively adjusts the global Lipschitz constant to reach this operating point. Through a margin parameter in the training loss, LiST further enables the construction of a fully calibrated Pareto front, allowing users to navigate the accuracy-robustness trade-off while remaining calibrated throughout. At convergence, LiST also enables the reintegration of calibration data into training, improving sample efficiency without sacrificing calibration. We validate LiST on CIFAR-10/100 and Tiny-ImageNet, demonstrating competitive accuracy and robustness against constrained and unconstrained baselines, while remaining calibrated out of the box. Code is available at GitHub.

AI 总结：探讨如何从组合优化视角实现可信机器学习，回顾并综合其与组合优化交叉领域进展，涵盖训练及训练后多任务，指出组合优化公式比纯启发式方法更具优势，虽有挑战，但在可信机器学习系统设计和部署中作用将增大。

链接：https://arxiv.org/abs/2607.07762

作者：Thibaut Vidal, Julien Ferry

英文摘要：Modern machine learning (ML) increasingly relies on complex models whose behavior is difficult to characterize beyond empirical performance metrics. Across a wide range of tasks, including prediction, generation, and decision-making, models with similar empirical performance can exhibit markedly different properties in terms of their transparency, interpretability, robustness, fairness, privacy, and certifiability. This survey highlights how optimization- and certification-oriented reasoning can provide a useful framework for reasoning about such differences, supporting tasks ranging from model training and selection to auditing and certification. We review and synthesize recent advances at the intersection of combinatorial optimization (CO) and trustworthy ML, covering both training and post-training tasks, including interpretable model learning, explanation generation, robustness analysis, fairness auditing, model compression, and privacy attacks and protections. Across these domains, CO formulations offer additional capabilities over purely heuristic approaches, e.g., gradient-based ones, notably global guarantees, formal certificates, and explicit treatment of trade-offs. While scalability remains an important challenge, continued progress in solvers and hybrid algorithms suggests a growing role for CO in the design and deployment of trustworthy ML systems.

AI 总结：研究针对大语言模型安全方法易受攻击问题，提出潜在人格对齐（LPA）方法，基于66条陈述训练，通过假设人格表征与避害共享结构，实现高攻击成功率、轻量级训练及良好泛化性。

链接：https://arxiv.org/abs/2607.07918

机构：Mila, Quebec AI Institute（米拉，魁北克人工智能研究所）； McGill University（麦吉尔大学）； LawZero； Université de Montréal（蒙特利尔大学）

作者：Mohamed Amine Merzouk, Nolan Smyth, Damiano Fornasiere, Linh Le, David Williams-King, Adam Oberman

英文摘要：Current safety methods for large language models are known to be vulnerable to adversarial attacks, motivating research into robust alternatives. Latent Adversarial Training (LAT) is among the most effective defenses, but can degrade utility and requires training on large datasets of harmful prompts. We introduce Latent Personality Alignment (LPA), which replaces explicit harm refusal with adversarial training on just 66 harm-agnostic statements drawn from psychometric personality literature. We hypothesize that personality-anchored representations share latent structure with harm avoidance, so adversarially stabilizing them implicitly constrains the subspace exploited by jailbreak attacks. LPA achieves near-zero attack success rates on HarmBench across direct requests and five jailbreak methods, despite never seeing harmful content during training and no loss of performance on standard benchmarks. Moreover, the training process is lightweight; the entire procedure completes in minutes on a single GPU and uses 75x fewer examples than standard LAT. Extensive ablations demonstrate the robustness, efficiency, and generalization of our method.

AI 总结：研究针对大语言模型辅助安全分析工具自身缺乏分析的问题，提出结构化元安全分析过程，通过运行元安全分析过程推导治理章程，包含21条工具原则和8条元安全原则，还报告了自推导等四个发现。

链接：https://arxiv.org/abs/2607.08054

作者：Samuel Tetteh, Udip Shrestha, Joshua R. Waite, Cody Fleming

英文摘要：Large language models (LLMs) are increasingly trusted to draft the artifacts of safety analysis such as, losses, hazards, Unsafe Control Actions (UCAs), and safety constraints, inside rigorous processes such as Systems-Theoretic Process Analysis (STPA). Yet a blind spot runs through this fast-growing literature: every system gets analysed except the LLM-assisted tool doing the analysing, which is itself a safety-relevant system that can hallucinate standards, emit unverifiable constraints, and leave no audit trail from prompt to artifact. We take seriously the question the field has skipped -- {who analyses the analyser?} and answer it by turning STPA on the tool itself. We present \{Constitutional Meta-STPA}, an LLM-assisted STPA tool built around a closed loop: the tool runs a {meta-STPA} of the class of AI-assisted safety tools and {derives} rather than asserts, its governance constitution from the resulting loss$\to$hazard$\to$UCA$\to$constraint chain, yielding a published constitution of $21$ Tool Principles and $8$ Meta-Safety Principles, each bound to a code enforcement point. We formalise the measured object as a constitution-marginal coverage operator over a principle set $P$ ($|P|{=}29$) with a soundness lemma that isolates coverage from model and scanner, and report four findings. {(i)~Self-derivation:} a frontier ensemble ({claude-opus-4.8}${+}${claude-sonnet-4}) recovers $18/21$ canonical and all $8/8$ governance principles from the tool's own design, while a weaker pair recovers $12/21$ and $3/8$, so the meta layer is model-limited, not constitution-limited, and the same $8/8$ re-emerge from a second, independently authored tool.

AI 总结：研究机制可解释性中因果声明评估问题，提出认证干预保真度（CIF）方法，通过将报告量写为因果估计量，提供置信区间等，能验证高保真声明，明确方法差异及对干预分布的敏感性，还降低了认证成本。

链接：https://arxiv.org/abs/2607.08349

机构：Vanderbilt University Medical Center（范德比尔特大学医学中心）

作者：Amir Asiaee

英文摘要：Mechanistic interpretability often evaluates explanations by intervening on a model: swapping hidden states, patching activations, ablating components, or comparing a compressed model to the original one. These experiments are usually summarized by a point estimate, even though the evaluation may be monitored while it runs or adapted toward suspected failures. This makes it hard to tell whether a reported fidelity or patching effect is a stable causal claim or a consequence of finite sampling and evaluation choices. We introduce Certified Interventional Fidelity (CIF), a statistical layer for interventional interpretability evaluations. CIF first writes the quantity being reported as a causal estimand: an expectation of a bounded score over a stated input distribution and a stated intervention distribution. It then provides confidence intervals and anytime-valid confidence sequences for this estimand, including under adaptive intervention sampling via bounded mixture importance weighting. We instantiate CIF with Hoeffding-style sequences and variance-adaptive betting sequences, the latter reducing certification cost by 10-30x in our experiments. On MNIST abstractions and GPT-2 Small IOI circuits, CIF certifies high-fidelity claims, shows when apparent method differences are not statistically supported, and makes sensitivity to the intervention distribution explicit.

AI 总结：针对复杂工业系统异常检测中常忽略内部因果关系的问题，提出CAAD框架，通过外生变量持续验证格兰杰因果一致性，利用多尺度对齐和梯度矩阵监测因果关系，在真实工业数据集实验中实现高精度异常检测，优于多数基线方法。

链接：https://arxiv.org/abs/2607.08555

机构：Stony Brook University（纽约州立大学石溪分校）； Rensselaer Polytechnic Institute（伦斯勒理工学院）； Yale University（耶鲁大学）； Queen Mary University of London（伦敦大学玛丽皇后学院）

作者：Xin Wang, Yunshi Wen, Yanan He, Haotian Xu, Youlan Zhao, Michel Ferreira Cardia Haddad, Tengfei Ma

英文摘要：The operational integrity of complex industrial systems relies on precise anomaly detection and diagnosis. The vast majority of existing methods narrowly focus on capturing temporal similarities of representations, often overlooking the disruption of internal causal relationships, which characterizes system failures and latent anomalies. In this paper, we propose a novel framework (CAAD) that reframes anomaly detection as the continuous verification of Granger causality consistency through exogenous variables. Specifically, the CAAD framework models exogenous time-series variables as residuals, identifying anomalies as significant deviations caused by external interventions. The proposed framework leverages multi-scale alignment to internalize system dynamics and utilizes a gradient-based matrix to monitor internal causal relationship breakdowns. By quantifying causal deviations of both dynamic evolution and relational topology, the CAAD is able to capture subtle causal shifts to achieve precise anomaly detection. Extensive experiments on real-world industrial datasets demonstrate that the CAAD achieves high-precision anomaly detection, outperforming most state-of-the-art baselines.

AI 总结：研究对抗性不确定性下的决策问题，基于贝叶斯决策理论，提出序列对抗鲁棒决策感知实验设计方法，通过形式化对抗鲁棒最优决策和推导设计标准，使决策在对抗变化下更稳定可靠。

链接：https://arxiv.org/abs/2607.08590

机构：The University of Manchester（曼彻斯特大学）； ELLIS Institute Finland（芬兰ELLIS研究所）； Aalto University（阿尔托大学）

作者：Haripriya Harikumar, Sammie Katt, Yasir Zubayr Barlas, Samuel Kaski

英文摘要： Scientific experiments are often designed to maximize information gain, yet in many applications the primary objective is to support reliable downstream decision-making. Existing decision-aware experimental design and active learning methods typically assume well-specified outcome models and implicitly rely on the stability of the optimal decision under real-world perturbations. In practice, however, experimental outcomes are frequently influenced by hidden or weakly modeled effects, which can substantially alter decision optimality and lead to misleading conclusions. We study sequential adversarially robust decision-aware experimental design, where data acquisition has to take into account information gain against plausible worst-case unexpected effects, modeled here as variation in adversarial variables. Building on Bayesian decision theory, we formalize an adversarially robust optimal decision under this setting and derive a principled Bayesian experimental design criterion. The criterion explicitly targets decision stability rather than nominal optimality. Experiments on synthetic and real-world scientific datasets show that conventional decision-aware design can converge rapidly to high confidence yet fragile decisions, while our robustness-aware approach yields decisions that are significantly more stable and reliable under adversarial variation.

AI 总结：研究如何增强时间图网络（TGNs）的可解释性，通过拓扑归因树和记忆回溯树对其预测进行归因，应用LRP并设计优化目标，在九个时间图数据集实验中表现出色，提供可靠解释且优于基线。

链接：https://arxiv.org/abs/2607.07716

作者：Yazheng Liu, Xi Zhang, Sihong Xie, Hui Xiong

英文摘要：Temporal graphs are ubiquitous in real-world applications and Temporal Graph Networks (TGNs) have achieved superior predictive accuracy. Understanding which historical events drive model predictions can enhance trustworthiness of TGNs. Existing explanation methods overlook the memory module, the core component that records and updates node histories, leaving the influence of past events unexplored. To address this, we attribute TGNs predictions through the topology attribution tree and memory backtracking tree. The topology attribution tree captures the influence of neighbors and their memory vectors, then the memory backtracking tree quantifies how historical events shape node memory vectors. We apply the LRP in TGNs, ensuring that the total contribution of events equals the logits of model. Finally, top-k selection may be unfaithful due to the nonlinear mapping from logits to probabilities, we design optimization objectives to identify the important events. Experiments on nine temporal graph datasets, spanning node property prediction, link prediction tasks and graph classification tasks, show that our method provides faithful explanations and outperforms state-of-the-art baselines. The code is available at this https URL

AI 总结：研究基于脑电图的情绪识别，提出图正则化学习框架，将情绪视为图中节点，采用三种正则化策略，在三种骨干架构上评估，在SEED-IV和SEED-V数据集上有改进，提升了标准方法的性能上限。

链接：https://arxiv.org/abs/2607.07773

机构：School of Mathematics (Zhuhai), Sun Yat-sen University（中山大学数学学院（珠海））

作者：Dongyang Kuang, Zizheng Ma, Yushan Zhang, Xiaocong Zeng

英文摘要：EEG-based emotion recognition is critical for mental health monitoring and affective brain-computer interfaces, yet existing deep learning approaches often treat emotion classes as isolated labels, ignoring their psychological interdependencies. We propose a graph-regularized learning framework that conceptualizes emotions as nodes in a graph where edges encode proximity based on dimensional emotion theories. We adapt three complementary regularization strategies--Graph Label Smoothing (intuitive soft labeling), Commuting distance on graph via Graph Laplacian (spectral graph theory), and Sliced Wasserstein Distance (optimal transport on graph)--ordered by increasing computational complexity. These strategies penalize model predictions that deviate from the established emotion topology. Our framework is evaluated across three representative backbone architectures: AudioTransformer (pure transformer), Conformer (CNN-transformer hybrid), and DCGNN (causal graph neural network), demonstrating architecture-agnostic benefits. Experiments on SEED-IV (4 classes) and SEED-V (5 classes) datasets show consistent improvements: best case up to +5.42% accuracy and 39% reduction in psychologically implausible misclassifications. Ultimately, our framework help raise the upper bound of performance achievable with standard approaches. Code will be released.

AI 总结：介绍用于图结构数据可解释监督学习的Python包path_boost，其实现的PathBoost算法能自动发现图中预测路径，基于路径特征生成可解释模型，避免穷举路径，支持多种功能，经分子性质预测实验并与其他方法对比，已开源。

链接：https://arxiv.org/abs/2607.07935

机构：University of Oslo（奥斯陆大学）

作者：Claudio Meggio, Johan Pensar, Riccardo De Bin

英文摘要：We present path_boost, a Python package for interpretable supervised learning on graph-structured input data. The package implements PathBoost, a gradient boosting algorithm that automatically discovers predictive labeled paths within graphs during the learning process. Unlike graph neural networks, which are generally difficult to interpret, PathBoost produces an additive prediction model over path-based features that explicitly reveals which substructures drive predictions. To avoid an exhaustive enumeration of all possible paths, the algorithm iteratively selects and extends paths during learning based on their predictive power, using boosting to combine weak learners into a strong ensemble. The package supports both regression and binary classification. Key features include compatibility with scikit-learn workflows, support for custom base learners and selectors, automatic starting node selection, parallel training across anchor nodes, and built-in variable importance computation. We demonstrate PathBoost on molecular property prediction of transition metal compounds, where atoms serve as nodes and bonds as edges, and further benchmark PathBoost against an established graph neural network and a graph kernel method across six molecular datasets. The package is available on PyPI and GitHub under an open-source license.

AI 总结： 针对现有技术无法用于聚类数据结构学习的问题，提出新方法，将经典混合模型框架扩展到结构学习，通过可微图耦合机制、收敛一阶方法等，在聚类数据中检测到其他估计器遗漏的依赖关系。

链接：https://arxiv.org/abs/2607.08238

机构：University of Technology Sydney（悉尼科技大学）； University of Michigan（密歇根大学）

作者：Ryan Thompson, Matt P. Wand, Veerabhadran Baladandayuthapani

英文摘要：Recent algorithmic advances have made directed acyclic graph (DAG) structure learning scalable for causal discovery. Yet, the currently available techniques assume a completely homogeneous population, precluding their application to clustered data where cluster-specific variations (e.g., patient-specific effects) are common. We address this issue by introducing a new approach that estimates a global structure while accounting for local cluster-level effects. The key idea is to extend the fixed- and random-effects framework of classical mixed models to the structure learning setting. Towards this end, we present a differentiable graph coupling mechanism that guarantees the union of the fixed- and random-effects graphs remains acyclic. Computationally, we provide a provably convergent first-order method and leverage efficient batched updates across clusters. Statistically, we establish identifiability of the model and show that our approach recovers the true structure asymptotically. In experiments on real and synthetic data, our proposal detects dependencies missed by alternative estimators, underscoring its value for structure learning in clustered settings.

AI 总结：研究针对持续语言模型微调中类似LoRA方法新任务覆盖旧任务的问题，提出ReCoLoRA框架，通过特定初始化、选秩及递归合并等操作，在多任务持续GLUE序列实验中，用更少参数取得更好成绩。

链接：https://arxiv.org/abs/2607.07719

作者：Wentao Lu

英文摘要：Parameter-efficient fine-tuning adapts a large language model to one task cheaply, but across a task sequence LoRA-style methods keep stacking low-rank updates on the same frozen weight, so each new task tends to overwrite the previous ones. We present ReCoLoRA (Recursive Consolidation of Low-Rank Adapters), a spectrum-aware framework for continual fine-tuning: adapters are initialized from a randomized SVD of the pretrained weight, per-layer effective ranks are selected by an elbow criterion, and the principal subspace is adapted before residual capacity is opened. Before each new task, ReCoLoRA re-decomposes the current effective weight, rather than the original one, into a frozen residual, a slowly updated principal component, and a fresh adapter (recursive consolidation), so every task starts from the model that has already absorbed its predecessors. On a six-task continual GLUE sequence over four 7-8B backbones, ReCoLoRA attains the best final average score on three of the four backbones against rank-swept LoRA, PiSSA, AdaLoRA, and DoRA baselines while training fewer parameters; an oracle-routed task-bank variant serves as an upper bound under full task isolation. Code: this https URL.

AI 总结：研究大语言模型持续学习问题，从空间和时间轴分解变化，通过重新设置基准和引入协议评估多种方法，发现持续学习非单一能力，不同环境变化需不同更新行为，为设计更强持续学习系统提供指导。

链接：https://arxiv.org/abs/2607.07847

机构：UC Berkeley（加州大学伯克利分校）； Capital Fund Management（资本基金管理公司）

作者：Anne Harrington, Nayan Saxena, Michael Murphy, Anastasia Borovykh, Zeyu Yun, Sridhar Kamath, Ara Eindra Kyi, Trevor Darrell, Jitendra Malik, Yutong Bai

英文摘要：As large language models (LLMs) become increasingly capable, the next question is how can we enable models to continually learn? Today, the field largely frames this as a problem of context management and mitigating forgetting. We argue this framing is incomplete: continual learning is fundamentally about increasing model competence as the world changes. We disentangle this change along two axes -- space, where the model encounters new domains, and time, where the underlying data drifts under a fixed task. This framing lets us study continual learning under realistic conditions: new domains arrive over time, facts drift past their training cutoff, and agentic interactions accumulate state across episodes. To evaluate methods under this setting, we recast widely used LLM benchmarks as sequential problems and introduce a single mechanism-agnostic protocol that compares prompt-based methods (GEPA, ACE), supervised learning (SFT, SDFT), reinforcement learning (GRPO, SDPO), and context compression (Cartridges, In-place TTT). Prompt-based methods fit each new stage quickly but degrade on future tasks. Distillation-based methods accumulate knowledge stably but struggle to update outdated facts. Context compression improves efficiency without substantially improving the ability to learn new tasks. Online reinforcement learning adapts most effectively to knowledge updates but remains sensitive to noisy reward signals. Overall, our results suggest that continual learning is not a single capability: different patterns of environmental change require fundamentally different update behaviors, determining when adaptation must be learned inside model weights and when it can be achieved through external scaffolding. We hope that understanding where each method succeeds and fails will guide the design of stronger continual learning systems.

AI 总结：研究首次评估表格模型作图像-表格设置的编码器，指出表格领域最佳的上下文学习模型因需标签处理实例，嵌入训练和测试实例不易，通过解决此问题强调了编码器因素在多模态学习中的重要性。

链接：https://arxiv.org/abs/2607.07756

作者：Ilia Koloiarov, Diego Coello de Portugal Mecke, Vijaya Krishna Yalavarthi, Tom Hanika, Lars Schmidt-Thieme

英文摘要： Multimodal learning usually requires a dedicated encoder per modality. When a tabular modality is involved, prior work has been mostly using a \emph{plain MLP} as the encoder. Yet if it were a strong encoder, the tabular domain would not be ``the last unconquered castle for deep learning''. This study evaluates state-of-the-art tabular models as encoders in the image-tabular setting for the first time. An obstacle stands out. In-Context Learning models, among the best performing methods in the tabular domain, require labels to process instances, making it non-trivial to embed training and test instances the same way. We addressed this problem across multiple models of this family. With this study, we would like to highlight the importance of encoder factor in the multimodal learning.

AI 总结：综述跨视觉、语言、音频和视频的多模态遗忘，基于相关进展等提供统一视角，通过分类法系统比较模型架构和模态，阐明权衡，强调开放问题与实际考虑，支持未来研究与部署，并发布存储库。

链接：https://arxiv.org/abs/2607.07907

作者：Nobin Sarwar, Shubhashis Roy Dipta, Zheyuan Liu, Vaidehi Patil

英文摘要：With the growing adoption of VLMs, DMs, LLMs, and AFMs, these multimodal foundation models can inadvertently encode sensitive, copyrighted, biased, or unsafe cross-modal associations that originate from their training data. Retraining after deletion requests or policy updates is often impractical, and targeted forgetting remains difficult because knowledge is distributed across shared representations. Multimodal unlearning addresses this challenge by enabling selective removal across modalities while retaining overall utility. This survey offers a unified, system-oriented view of multimodal unlearning across vision, language, audio, and video, grounded in recent advances, emerging applications, and open problems. Our taxonomy enables systematic comparison across model architectures and modalities, clarifying trade-offs among deletion strength, retention, efficiency, reversibility, and robustness. This survey highlights open problems and practical considerations to support future research and deployment of multimodal unlearning. We release a curated repository: this https URL

AI 总结：研究以加利福尼亚野火PM2.5为例，比较六种时间序列基础模型配置与全训练基线及朴素持久性方法，通过留一事件法评估泛化能力，发现BiLSTM表现最佳，零样本TSFMs改进有限，LoRA微调有改善但未超循环基线，挑战大预训练模型主导环境预测的假设并提供指导。

链接：https://arxiv.org/abs/2607.07951

机构：College of Engineering, University of Georgia（佐治亚大学工程学院）； Information System Department, UMBC（马里兰大学巴尔的摩郡分校信息系统系）； College of Graduate and Professional Studies, Trine University（特瑞大学研究生与专业研究学院）； School of Professional Studies, Columbia University（哥伦比亚大学专业研究学院）

作者：Yongcan Huang, Li Jiang, Ze Yu Liu

英文摘要：Wildfire smoke events produce extreme PM$_{2.5}$ concentrations that pose severe public health risks, yet forecasting rare, hazardous-level spikes remains a fundamental challenge. Time series foundation models (TSFMs), pretrained models offering zero-shot inference and efficient adaptation, perform strongly on general benchmarks, but their behavior under extreme out-of-distribution conditions is poorly understood. We present the first systematic benchmark comparing six TSFM configurations (zero-shot TimesFM, Chronos-2, Moirai-2, and Time-MoE, plus LoRA fine-tuned Chronos-2 and Time-MoE) against fully-trained baselines (LSTM, BiLSTM, Transformer) and naive persistence on a 12-year (2013--2025) hourly PM$_{2.5}$ dataset covering 1,375 wildfire incidents across 79 California monitoring sites. A leave-one-incident-out (LOIO) protocol evaluates generalization to unseen fires, using MAE, RMSE, and exceedance F1 at EPA AQI thresholds across 6-, 12-, and 24-hour horizons. Results reveal a consistent hierarchy. The BiLSTM achieves the lowest MAE ($5.16\,\mu g/m^3$) and the highest exceedance F1 at every threshold, including the Hazardous band ($>225.5\,\mu g/m^3$), reaching 0.63 versus at most 0.54 for any foundation model. Zero-shot TSFMs improve on persistence only modestly, and zero-shot Chronos-2 exhibits severe RMSE tail instability ($23.4\,\mu g/m^3$, negative $R^2$) from sporadic large errors. LoRA fine-tuning substantially improves both adapted families and largely repairs this instability, yet no foundation model surpasses the trained recurrent baselines on any metric. These findings challenge the assumption that larger pretrained models universally dominate environmental forecasting and provide actionable deployment guidance for wildfire air quality prediction.

AI 总结：研究指出固定大小基准用于模型评估效率低，提出自适应评估框架，结合序贯测试统计范式与定制停止标准，在Open VLM Leaderboard上展示能自适应管理效率与可靠性权衡，降低计算成本并保持统计显著性。

链接：https://arxiv.org/abs/2607.08522

机构：IBM Research（IBM研究院）

作者：Ofir Arviv, Kristjan Greenewald, Yotam Perlitz, Hadar Mulian, Michal Shmueli-Scheuer, Leshem Choshen

英文摘要：The inherent rigidity of fixed-size benchmarks makes them an inefficient tool for model evaluation. Diverse evaluation objectives, including model ranking, model selection and testing throughout development, demand varying levels of statistical power. The mismatch between fixed sample sizes and these diverse needs results in either excessive computational cost or compromised reliability - a critical concern for model evaluation. To overcome these limitations, we call for adoption of sequential testing in our field. We provide an adaptive evaluation framework, that provides a principled way to navigate the trade-off between efficiency and reliability in model evaluation. Our framework combines the established statistical paradigm of sequential testing with stopping criteria tailored to common evaluation needs such as diminishing returns detection, and minimum detectable effect size. We demonstrate its ability to adaptively manage the efficiency-reliability trade-off on the Open VLM Leaderboard, including, for example, a 80% reduction in computational cost compared to fixed-size evaluation (with a 2.5-point CI width allowance) while maintaining statistical significance.

AI 总结：研究胸部X光分类中长尾数据下子组诊断不足问题，通过诊断阶梯分离相关因素，在VinDr-CXR和MIMIC-CXR/CXR-LT数据集上实验，表明CXR罕见标签公平性取决于发现、子组和阈值，非仅标签频率或排序指标。

链接：https://arxiv.org/abs/2607.07717

机构：University of Science, Ho Chi Minh City（胡志明市科学大学）； Vietnam National University, Ho Chi Minh City（胡志明市越南国立大学）； VinUni-Illinois Smart Health Center, VinUniversity（VinUni-伊利诺伊智能健康中心，Vin大学）； Carnegie Mellon University（卡内基梅隆大学）； Northwestern University（西北大学）； College of Engineering & Computer Science, VinUniversity（Vin大学工程与计算机科学学院）

作者：Ha-Hieu Pham, Hai-Dang Nguyen, Dang P. M. Cao, Thanh-Huy Nguyen, Min Xu, Trung-Nghia Le, Ulas Bagci, Huy-Hieu Pham

英文摘要：In chest X-ray (CXR) classification, acceptable ranking performance can still leave rare-positive patients below threshold, especially within subgroups. We study this pre-deployment fairness problem as an audit question: after a long-tailed multi-label CXR model is converted from scores into decisions, who is missed? Across VinDr-CXR and MIMIC-CXR/CXR-LT, we use a diagnostic ladder to separate class-level long-tail losses, subgroup-aware weighting, group robustness, and threshold selection. On VinDr-CXR, group-tail weighting followed by tail-aware thresholding reduces tail FNR from 0.665 to 0.269, sex worst-group FNR from 0.705 to 0.157, and age worst-group FNR from 0.822 to 0.133, while macro-mAP increases from 0.611 to 0.635. On MIMIC-CXR/CXR-LT, the same score-to-threshold comparison reduces tail FNR from 0.866 to 0.741 and lowers worst-group FNR across sex, age, race, and insurance; residual missed-positive rates nevertheless remain high. Paired bootstrap contrasts on VinDr support the thresholded FNR reductions, and GroupDRO reference runs indicate that aggregate group robustness alone does not remove rare subgroup misses in this setting. The study supports a narrow audit claim: rare-label fairness in CXR depends jointly on the finding, subgroup, and operating threshold, not on label frequency or ranking metrics alone.

AI 总结：研究针对基因组预测模型跨机构转移问题，提出SHIFT模型，通过分别表示基因组特征、利用掩码自注意力等方法，直接从不完整输入预测，在多队列实验中展现强泛化能力且优于其他方法，支持缺失感知建模用于多中心生存预测。

链接：https://arxiv.org/abs/2607.07725

作者：Muhammet Sami Yavuz, Ayhan Can Erdur, Sabri Mustafa Kahya, Benedikt Wiestler, Jana Lipkova

英文摘要：Genomic prediction models often fail to transfer across institutions because sequencing panels differ across sites, creating structural feature missingness at deployment. Existing approaches to this challenge typically restrict analysis to genes shared across cohorts, exclude patients with incomplete profiles, or rely on test-time imputation, all of which can reduce robustness and limit the use of multi-center data. We propose Survival prediction Handling Incomplete Features using Transformer (SHIFT), a missingness-aware survival model that directly predicts from incomplete genomic inputs without test-time imputation. SHIFT represents each genomic feature separately and uses masked self-attention, along with a feature-availability mask, so that predictions are based only on observed inputs. Further, we introduce variable-rate feature masking during training to improve robustness to heterogeneous missingness patterns. We evaluate the approach on glioblastoma and lung squamous cell carcinoma with external validation across multiple cohorts, including a challenging setting with severe cross-cohort panel mismatch. Across these settings, SHIFT shows strong generalization and compares favorably with standard survival baselines and imputation-based approaches, while using a single model across differing feature sets. We also find that incorporating patients from incomplete cohorts during development can improve performance on external data, suggesting that partially observed cohorts need not be excluded from model building. These results support missingness-aware modeling as a practical strategy for multi-center survival prediction in precision oncology.

AI 总结：研究地球观测基础模型，指出其因数据特性需特定领域适配。回顾相关设计原则，综合模型格局等内容，纳入基准证据说明评估问题，通过案例展示实践原则，提出应从多方面评估下一代遥感基础模型。

链接：https://arxiv.org/abs/2607.07758

作者：Syed Usama Imtiaz, Mitra Nasr Azadani, Nasrin Alamdari

英文摘要：Foundation models (FMs) have transformed machine learning from isolated task-specific model development toward general-purpose models pretrained on broad data and adapted to multiple downstream tasks. Earth observation (EO) is an important domain for this paradigm because satellite and airborne archives are large, high-revisit, and increasingly multimodal, while reliable field labels are often sparse. Remote sensing foundation models (RSFMs) cannot be transferred reliably/optimally without domain-specific adaptation. This is because EO data are governed by measurement physics and operational decision constraints. This chapter reviews the design principles arising from these domain-specific constraints. It first defines the FMs paradigm in remote sensing (RS), then synthesizes the current model landscape, pretraining objectives, architecture designs, downstream adaptation and trustworthiness requirements. The chapter also incorporates recent benchmark evidence showing that no single geospatial foundation model is universally best and that inconsistent evaluation remains a major issue to fair comparison and reliable deployment. In addition, two brief environmental monitoring case studies; physics-informed spectral targeted masking for harmful algal bloom prediction and reinforcement learning for adaptive environmental monitoring station selection to illustrate the FMs domain-guided principles in practice. This chapter posits that next-generation RSFMs should be evaluated not only by benchmark accuracy, but also by modality-aware transfer and physically plausible representations for trustworthy EO decisions.

AI 总结：研究利用卫星衍生预测因子，开发时空机器学习框架预测葡萄牙海岸拟菱形藻有害藻华。通过对贝类生产区观测，针对热点区域建模，采用特殊聚类和交叉验证策略，结果显示有一定可预测性，集成树方法效果好，框架对HAB预警系统有操作相关性。

链接：https://arxiv.org/abs/2607.07834

作者：Ayman Bnoussaad, El Khalil Cherif, Ligia Pinto, Ramiro Neves, Alexandra D. Silva, Alexandre Bernardino

英文摘要： Pseudo-nitzschia diatoms pose recurrent risks to coastal ecosystems and shellfish harvesting along the Portuguese Atlantic coast. Here we develop and evaluate a spatio-temporal machine-learning framework to predict harmful algal bloom (HAB) occurrence using exclusively satellite-derived predictors under realistic forecasting constraints. We characterised environmental and biological variability across shellfish production zones (L1-L9) using 5,882 observations, providing system-wide context. Predictive models were developed for zones L1-L2, a hotspot for Pseudo-nitzschia and domoic acid events, using a decade-long dataset (2013-2023; 1,440 observations; more than 1,000 satellite-based predictors including sea surface temperature, an upwelling index, chlorophyll-a, and plankton functional types). Sampling locations were partitioned into ecologically meaningful sub-regions using a river-aware spatial clustering scheme. A stringent spatio-temporal cross-validation strategy that simultaneously withholds entire years and spatial clusters prevents leakage and closely mimics real-world forecasting conditions. HAB occurrence proved moderately predictable across model classes and feature configurations. Ensemble tree-based methods achieved the strongest discrimination: Random Forest reached 0.74 +/- 0.05 with environmental predictors; Extra Trees reached 0.77 +/- 0.06 with biological variables added. Feature-importance analyses revealed that seasonal structure, spatial context, and lagged environmental conditions dominate model decisions, while biological indicators refine bloom likelihood within physically favourable periods. The framework demonstrates operationally relevant skill for satellite-supported HAB early-warning systems along eastern boundary upwelling coasts.

AI 总结：针对符号回归面临的高维输入等挑战，DeepPySR通过动态变量剪枝、指数帕累托选择标准和分层符号组合架构来解决，在多个数据集上表现出色，能产生与领域风险因素一致的可解释公式。

链接：https://arxiv.org/abs/2607.08150

机构：International Centre for Radio Astronomy Research (ICRAR), University of Western Australia（国际射电天文学研究中心（ICRAR），西澳大利亚大学）； The Raine Study, University of Western Australia（雷恩研究中心，西澳大利亚大学）； Menzies Institute for Medical Research, University of Tasmania（孟席斯医学研究所，塔斯马尼亚大学）； Nutrition & Health Innovation Research Institute, School of Medical and Health Sciences, Edith Cowan University（营养与健康创新研究所，伊迪斯科文大学医学与健康科学学院）

作者：Fuling Chen, Kevin Vinsen, Phillip Melton, Rae-Chi Huang

英文摘要：Symbolic regression (SR) discovers analytical equations from data, yielding glass-box models with directly interpretable formulas, unlike black-box methods that rely on unstable post-hoc tools such as SHAP or LIME. This transparency is crucial in clinical medicine and social science, but SR faces three challenges: high-dimensional inputs, principled selection of Pareto-front formulae, and data irregularities such as multicollinearity and class imbalance. We introduce DeepPySR, which addresses these issues with a dynamic variable-pruning schedule to remove irrelevant features during search, an exponential Pareto selection criterion that eliminates trade-offs between accuracy and complexity, and a multi-layer architecture for hierarchical symbolic composition. On four Feynman physics benchmarks and seven biomedical and social-science datasets, DeepPySR outperforms PySR and baselines on body fat (R$^2$: 0.794 vs.\ 0.702), heart disease (F1: 0.898 vs.\ 0.787), student performance (R$^2$: 0.964 vs.\ 0.948), and Raine BMI (R$^2$: 0.525 vs.\ 0.370), producing interpretable formulas aligned with domain risk factors.

AI 总结：研究针对病理检查推荐延迟问题，引入基于分类器链技术的系统，将其构建为多标签分类问题。收集数据应用多种算法比较模型，通过可解释人工智能技术确保模型透明度和临床可解释性，提高传统算法在诊断过程中的效率并提供准确推荐。

链接：https://arxiv.org/abs/2607.08299

作者：Abu Rafe Md Jamil, Nayan Malakar

英文摘要：Accurate and timely diagnoses are essential for quality patient care. However, delayed recommendation of diagnostic tests and physicians' subjective interpretations can hinder effective care. This study introduces a pathological test recommendation system that speeds up the test selection process using patient symptoms before physician consultation. The recommendation task is framed as a multi-label classification problem utilising the Classifier Chain (CC) technique to consider dependencies between tests. We collected data from the this http URL pathology and then created a custom dataset with the help of the expertise. Multiple machine learning algorithms, including Logistic Regression, Decision Tree, and Random Forest, were applied to compare models and identify the best fit for our study context. The Logistic Regression with CC model had the highest overall accuracy at 98.83%, while the Majority Voting ensemble model provided the best balance with a precision of 0.93, recall of 0.85, and F1-score of 0.89. To ensure transparency of the models and clinical interpretability, we used Explainable AI (XAI) techniques utilising SHAP (SHapley Additive Explanations), which identifies how each symptom is contributing to a test recommendation. The diagnostic reasoning revealed by the model was consistent with established medical knowledge of symptoms for the recommended tests, which further adds confidence to the model's reliability for diagnostic purposes. The reasoning could help physicians make logical decisions in critical scenarios. Overall, our findings suggest that CC can improve the efficiency of the traditional algorithms in diagnostic process providing accurate test recommendations.

AI 总结：研究利用VISEM数据集，基于精子浓度、活力和形态等精液参数，用机器学习算法分类男性生育状况，经预处理和特征工程，用LazyPredict框架训练评估模型，最近质心分类器准确率达94.2%，验证了模型稳健性，强调机器学习对生育诊断的潜力。

链接：https://arxiv.org/abs/2607.08429

机构：Pak-Austria Fachhochschule: Institute of Applied Sciences and Technology（巴基斯坦-奥地利应用科学与技术高等学院）； Mid Sweden University（中瑞典大学）； Coventry University（考文垂大学）； Saudi Standards, Metrology and Quality Organization（沙特标准、计量与质量组织）； Brightware LLC（Brightware有限责任公司）

作者：Shahnawaz Qureshi, Raja Khurram Shahzad, Muhammad Fozan, Emal Kawal, Syed Aziz Shah, Sattam Al-Anazi, Syed MuhammadZeeshan Iqbal

英文摘要： Male infertility is a significant yet often underdiagnosed aspect of reproductive health, with semen analysis serving as the cornerstone of clinical evaluation. To address this problem, this study investigates the use of machine learning algorithms to classify male fertility status based on key semen parameters, i.e., sperm concentration, motility, and morphology, using the VISEM dataset. This dataset includes semen samples from 85 participants, classified into three categories, i.e., Fertile, Sub-Fertile, and Infertile, according to the World Health Organization's criteria. After pre-processing and feature engineering, the dataset was used to train and assess multiple classification models using the LazyPredict framework. Among the more than 40 algorithms tested, the Nearest Centroid classifier achieved an accuracy of 94.2%, outperforming other models such as Support Vector Machines and Quadratic Discriminant Analysis. The model's robustness was validated using 5-fold cross-validation and multiclass ROC-AUC analysis. This study illustrates that machine learning models can provide fast, accurate, and objective assessments of semen quality, potentially supporting clinical decision-making in andrology and assisted reproductive technologies. These findings emphasize the growing potential of machine learning to enhance fertility diagnostics and inform patient-specific treatment strategies.

AI 总结：研究小数据约束下物理信息机器学习，以磨料水射流铣削数据集为例，做出三点方法学贡献，包括区分数据处理方式、发现模型排名不稳定及研究物理集成水平谱，为小且昂贵的过程数据集的可靠模型比较提供方法建议。

链接：https://arxiv.org/abs/2607.07863

作者：Sarah Grewe, Jörg Frochte

英文摘要：In physically dominated machining processes, experimental datasets are small, expensive, and material-specific; in this regime, data curation, evaluation design, and the form of physics integration can matter as much as the learning algorithm. Using an abrasive waterjet milling dataset ($n{=}155$, Inconel\,718), we make three methodological contributions. First, we separate physics-based data \emph{cleaning} from statistical \emph{curation} and treat the latter as competing modelling hypotheses rather than silent preprocessing. Second, we find that model rankings from a 15-point hold-out set can be unstable: the single-split winner drops from rank~1 to rank~7 under 10-fold cross-validation, while Gaussian Process (GP) variants occupy the top ranks. Third, we study a spectrum of physics integration levels and find that residual learning on a compact physics baseline is competitive for GP, yielding lower variance and an interpretable decomposition, but degrades tree-based models. Bayesian hyper parameter tuning improves parameter-sensitive baselines such as gradient boosting and SVR, yet harms multi-stage hybrid pipelines at this sample size. GP uncertainty intervals are approximately calibrated ($86\%$ empirical coverage at nominal $90\%$). The resulting picture is methodological: for small, expensive process datasets, our results suggest that, in this setting, reliable model comparison benefits from explicit curation hypotheses, robust evaluation, and careful choices about how physics enters the model.

AI 总结：研究OLS回归的分布式草图绘制，通过在分区子集上进行草图绘制降低计算成本，在固定设计设置下刻画平均OLS估计器的确切超额损失，结果表明子集协方差差异小时该损失与全数据集草图绘制的损失相当。

链接：https://arxiv.org/abs/2607.07888

机构：School of Computer Science, University of Oklahoma（俄克拉荷马大学计算机科学学院）

作者：Luyuan Yang, Brayden Garner, Shayan Shafaei, Chao Lan

英文摘要：This paper studies distributed sketching for ordinary least squares (OLS) regression, an approach that distributes small sketches of a large data set over multiple machines to separately construct OLS estimators and average them. Unlike prior studies that consider sketching on the whole data set, we consider sketching on partitioned subsets to further reduce computational cost. Under the fixed design setting, we characterize the exact excess loss of the averaged OLS estimator. Results show that this loss is comparable to the established loss for sketching on the whole data set when the divergence among subset covariances is small.

AI 总结：研究大语言模型和智能体内存压缩问题，提出基于率失真视角统一分类各层方法，构建七轴分类法并转移机制，还针对相关问题未被测量及缺乏统一基准的情况，转化为提案、实验、原则并梳理开放问题。

链接：https://arxiv.org/abs/2607.08032

作者：Ashwin Gerard Colaco, Nada Lahjouji

英文摘要：Large language models, and the agents built on them, spend an ever-growing share of their compute and memory on remembering: caching attention keys and values, carrying long prompts, maintaining recurrent state, and storing what happened in previous turns and sessions. Because none of this memory is free, four largely separate research communities have each learned to compact it. They evict or quantize the KV cache, prune or distill prompts, bound architectural state, and consolidate agent memory. We argue that these are instances of one problem: a rate--distortion decision about what context-derived information to retain versus discard, at what fidelity, under a resource budget, so as to preserve downstream task utility. We make this lens precise with a single compaction objective and a layer-agnostic lower bound, use it to build a seven-axis taxonomy that classifies methods from across the stack uniformly, and use it to transfer mechanisms between layers that have never been connected, from serving-stack KV management to agent long-term memory. Two patterns hold across the survey. At every layer the signal that decides what to keep is attention magnitude or recency, and it fails in the same way everywhere, by discarding, before the query is known and with no way to undo it, information the query later needs. And while compression is measured carefully on single-turn long context, the repeated compaction that agents actually perform is almost never measured, and no benchmark holds one budget axis across all the layers at once. We turn both observations into a benchmark proposal, a small reference experiment, and a set of compaction-aware design principles, and we map the open problems.

AI 总结：研究视觉语言模型推理链中的认知信号，通过在相同对抗样本上运行四个模型，发现不同模型的答案熵行为模式有差异，思考链熵在部分情况下优于答案熵，还发现结构化弃权及其实用弃权门可提升准确率。

链接：https://arxiv.org/abs/2607.08059

作者：Mayank Singal

英文摘要： Uncertainty quantification for visual language models (VLMs) conventionally targets the answer token distribution. We provide the first three-family empirical characterisation of answer entropy behaviour in thinking-mode VLMs. Running four models on identical POPE adversarial samples, we find three qualitatively distinct patterns: Qwen3-VL-8B-Thinking shows complete collapse (ans H AUROC = 0.492); GLM-4.1V-9B-Thinking shows no collapse (0.716); and InternVL3-8B shows selective thinking (chains on only 50% of samples, ans H = 0.675 full / 0.602 thinking-only). Across all three thinking-mode models, thinking chain entropy outperforms answer entropy on the subset where chains are generated (0.647, 0.759, 0.608 vs. 0.492, 0.716, 0.602 respectively), suggesting chain signals are the more reliable predictor whenever chains are present. This holds strongly for Qwen and GLM, but with only marginal and statistically unreliable advantage for InternVL3 (n_FP = 17). A 300-sample VQAv2 pilot confirms chain entropy (0.680) outperforms answer entropy (0.595) on VQAv2 questions, with the gap largest for free-form answers (0.733 vs. 0.467). On harder reasoning tasks (HallusionBench) both Qwen models show moderate signal (approx. 0.64), consistent with incomplete pre-commitment on difficult questions. We additionally document structured abstention affecting 12-22% of queries with asymmetry toward absent-object queries, and a practical abstention gate raising accuracy from 71.0% to 93.8% at 62.7% coverage with no additional inference cost.

AI 总结：研究如何从异构历史证据发现并形式化艺术创作过程，核心方法是将证据合成存储库，由溯因智能体推断生产步骤并优化，主要贡献是迈向以过程为中心的人机协同创作系统，支持多方面研究。

链接：https://arxiv.org/abs/2607.08331

作者：Kaustubh Kumar, Ashutosh Ranjan, Vivek Srivastava, Blessin Varkey, Shirish Karande

英文摘要：Understanding how artworks are created requires reasoning about the iterative decisions, material operations, and contextual influences that shape artistic production. While recent generative AI systems can synthesize artworks with high fidelity, they primarily model distributions over finished artifacts rather than the creative processes underlying their creation. In practice, artistic workflows are only partially documented through fragmented sources such as archival records, preparatory studies, correspondence, etc., making process-level understanding difficult to formalize computationally. In this work, we introduce ArtMine, a framework for discovering and formalizing artistic processes from heterogeneous historical evidence. Our approach synthesizes heterogeneous artwork evidence into a structured repository, from which a Peircean abductive agent infers evidence-grounded production steps. These steps are converted into a compositional graph and rendering prompt, then optimized through self-reflection over deviations between the generated and reference artworks. We provide a preliminary proof-of-concept case study using open-domain historical sources across multiple artists and artistic movements, demonstrating that fragmented documentary evidence can support coherent, interpretable, and auditable representations of artistic workflows. By modeling creative processes rather than only final artifacts, our work moves toward process-centred human-AI co-creativity systems that can support artistic interpretation, creative education, reflective collaboration, and computational studies of cultural production.

AI 总结：研究神经人工智能中深度神经网络模型与大脑比较及趋同进化问题，通过证明两个最小DNN解在硬任务中的对齐特性，形式化逆变概念，指出强任务下网络比较度量选择不敏感且趋同进化或不可避免。

链接：https://arxiv.org/abs/2607.08561

作者：Dan Yamins, Aran Nayebi

英文摘要：A series of results from the NeuroAI over the past fifteen years have raised core questions both about how to compare Deep Neural Network (DNN) models to the brain, and about how much convergent evolution to expect between artificial networks and real brain networks. Here, we show that for any two minimal DNN solutions to a sufficiently hard task: (i) "weak" alignment of network representations based on affine mappings guarantees "strong" alignment of privileged axes, and (ii) alignment "zippers" up the network hierarchy, causing the emergence of privileged axes from end-to-end task optimization. These results formalize the notion of contravariance from Cao and Yamins [2024], and illustrate important consequences for the theory of NeuroAI: with sufficiently strong tasks, choice of metric for inter-network comparison is not all that sensitive, and that convergent evolution is probably inevitable.

AI 总结：该研究针对机器学习模型可解释性问题，提出基于部分依赖关系引导神经网络训练的新方法，使模型对特定特征的平均响应符合领域知识。经实验验证，此方法在回归问题上表现更优且数据效率更高，解释也与用户知识相符。

链接：https://arxiv.org/abs/2607.08641

作者：Yann Claes, Pierre Geurts, Vân Anh Huynh-Thu

英文摘要：Over the last few years, there has been an increased interest in making machine learning models more interpretable. Although a great deal of effort goes into developing techniques for interpreting the interactions learned by a given model, fewer studies focus on assessing the quality of such explanations. Even fewer focus on how to adjust the model to produce explanations faithful to prior knowledge, a process known as explanation-guided learning. Furthermore, most approaches in this area focus on classification problems and usually assume prior knowledge about which input features or regions are most important. In this work, we introduce a new approach to steering neural networks based on partial dependence, such that their average response to certain features aligns with specific functional domain knowledge about the problem. We empirically demonstrate on a range of regression problems, including dynamical systems forecasting, that models whose training has been controlled using our method perform better than unconstrained models and are more data-efficient. Moreover, we highlight that interpretations obtained from the former actually align with the user-provided knowledge, whereas those obtained from the latter do not.

AI 总结：研究大语言模型测试时模型选择问题，提出预算感知测试时模型选择方法，通过在重采样和重新路由间分配预算单位最大化预期正确性，提出RoR分配策略，实验表明该策略在成本-质量方面优于多种基线。

链接：https://arxiv.org/abs/2607.08665

机构：Institute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung University（国立阳明交通大学生物信息与系统生物学研究所）； Krixvon（克瑞克斯冯）

作者： Teng-Ruei Chen

英文摘要：Routing among large language models (LLMs) trades response quality against serving cost, motivated by the reported gap between deployed routers and a per-instance oracle. Recent analysis shows that test-time resampling can recover per-instance selection headroom that no single-commit router captures; however, that guarantee holds only under an idealized oracle equipped with correctness labels and an unconstrained budget, neither of which a deployed system has. To the best of our knowledge, no previous work treats resampling the committed model and rerouting to an alternative model as competing uses of a single per-query cost budget. Therefore, this work formulates budget-aware test-time model selection: given a per-query budget and an imperfect verifier, allocate each unit of budget between resampling and rerouting so that expected correctness is maximized. An online resample-or-reroute (RoR) allocation policy driven by estimated marginal correctness per unit cost is proposed, and its behavior is grounded in the recoverability asymmetry between selection and sampling. Replay experiments on newly regenerated multi-draw correctness tensors from an eleven-model open-weight pool over four benchmarks of differing difficulty show that the proposed RoR policy attains a favorable cost-quality Pareto front relative to single-route, one-commit-router, budget-aware best-of-K, cascade, and random-allocation baselines for the tested pools, with the largest gains on the most heterogeneous benchmark; an ablation further shows the gains are verifier-gated, shrinking as verifier quality degrades, and robustness replays under a provider price vector and a label-free agreement verifier delineate where the conclusions carry over.

AI 总结：研究无训练的松弛推测解码技术，统一现有方法于共享框架，在当代环境下基准测试，发现松弛需大量能力评估，且许多方法依赖良好语言模型作为起草器，不适用于轻量级专用多令牌预测起草器。

链接：https://arxiv.org/abs/2607.08690

机构：Graphcore（Graphcore公司）

作者：Guoxuan Xia, Luka Ribar, Paul Balanca

英文摘要：Speculative decoding accelerates sampling from an autoregressive LLM by using a faster auxiliary model to draft tokens which are then verified in parallel by the LLM. Standard speculative decoding is lossless: its rejection and resampling steps exactly preserve the LLM's sampling distribution. Recent work argues that relaxing this strict guarantee can yield further speed-ups, controlled capability-speed trade-offs, or even capability gains. We practically investigate training-free relaxed speculative decoding techniques, unify existing approaches within a shared framework, benchmark them on contemporary settings, and distil takeaways and empirical findings for practitioners. Important takeaways include: relaxation can require considerable capability evaluation unlike lossless speculative decoding, and many relaxed approaches rely on a drafter that is a good language model, making them unsuited for lightweight dedicated multi-token-prediction drafters.

AI 总结：研究OFDM系统中NBI干扰问题，提出统一深度学习框架，包含NBI-CNet估计参数消除干扰及LLR-CNet映射残差，相比传统算法复杂度降低，能消除误码平台，在不同干扰场景有良好性能且可稳健泛化。

链接：https://arxiv.org/abs/2607.08717

作者：Emmanouil Kavvousanos, Francky Catthoor, Vassilis Paliouras

英文摘要：Narrowband interference (NBI) severely degrades orthogonal frequency-division multiplexing (OFDM) systems by corrupting subcarriers and rendering classical soft demodulation ineffective. Conventional compressed-sensing (CS) mitigation exhibits high sequential latency and leaves structured, non-Gaussian residuals that cause log-likelihood ratio (LLR) unreliability, decoder saturation, and severe error floors when employing classical Gaussian demappers. We resolve this pipeline mismatch using a unified deep learning framework for joint NBI cancellation and robust soft demodulation. First, NBI-CNet employs a physics-informed convolutional architecture to estimate NBI parameters and remove multi-tone interference in a single forward pass. Without requiring prior knowledge of the active interferer count, NBI-CNet reduces computational complexity by up to 60% ($N{=}2048, Q{=}64$) compared to the state-of-the-art EOMP-IDS algorithm. Second, LLR-CNet acts as a structural whitener by mapping non-Gaussian post-mitigation residuals onto well-calibrated soft metrics. Simulations demonstrate that this joint framework eliminates the error floors inherent to traditional baselines across dense grids. Under severe interference ($\text{SIR}{=}{-}10$ dB), the pipeline operates within a $0.2$ to $0.5$ dB SNR margin of the optimal iterative baseline at a target block error rate (BLER) of $10^{-4}$. Under mild interference ($\text{SIR}{=}10$ dB) with heavy spectral overlap ($Q{=}12$), where classical greedy algorithms erroneously subtract valid data components and corrupt the payload, NBI-CNet avoids signal-peak confusion to deliver a coding gain exceeding $3$ dB. Finally, the architecture circumvents the $2{\times}10^{-4}$ error floor triggered by interferer-estimation errors, while its scale-invariant design enables robust generalization across arbitrary FFT sizes without retraining.

AI 总结：研究大语言模型中超级权重，发现修剪超级权重致性能下降非普遍适用，孤立训练超级权重失败，普通LoRA更新少量参数成功，表明参数重要性不意味可孤立训练，有效微调依赖层的结构化分解。

链接：https://arxiv.org/abs/2607.08733

机构：Amazon Web Services（亚马逊网络服务公司）

作者：Shreyas Subramanian, Adewale Akinfaderin, Akarsha Sehwag

英文摘要： Recent work identified Super Weights, individual parameters whose removal degrades model performance by orders of magnitude. We show that this degradation due to pruning Super Weights does not universally apply to all LLMs. Furthermore, if these parameters are so important, Super Weight-aware training should be effective. We show the opposite. Training Super Weights in isolation (100 to 8,192 parameters) drops accuracy to random-guessing levels on both OLMo-1B and OLMo-7B, and expanding to local neighborhoods of up to 36K parameters provides no improvement. The failure is specific to Super Weight coordinates: training an equal number of randomly chosen positions in the same down_proj layers instead improves over the baseline, so the collapse comes from targeting Super Weights, not from sparsity itself. Vanilla LoRA, updating every position in attention weight matrices through low-rank structure, succeeds with only 0.16% of parameters, and applying the same low-rank update to down_proj succeeds as well. A 10-seed ablation confirms that constraining LoRA updates at positions corresponding to Super Weight coordinates yields statistically indistinguishable results. These findings establish that parameter importance does not imply parameter trainability in isolation, and that effective fine-tuning relies on structured decompositions over entire layers rather than targeting individually important weights.

AI 总结：研究探索UMAP内部kNN图的潜力，应用PageRank、k核分解和聚类系数等标准图算法增强数据意义建构，经对MNIST和Fashion MNIST评估，证明这些基于图的分析实用且与专门方法有竞争力或互补。

链接：https://arxiv.org/abs/2607.08746

机构：Apple（苹果公司）

作者：Duen Horng Chau, Donghao Ren, Fred Hohman, Dominik Moritz

英文摘要：While UMAP is widely used for exploring high-dimensional data, typical workflows focus on its lower-dimensional embedding, largely overlooking the rich k-nearest-neighbor (kNN) graph that UMAP constructs internally. This graph encodes the data manifold in its original high-dimensional space, before the distortion that UMAP's 2D projection introduces. We demonstrate the untapped potential of this internal representation, showing how standard graph algorithms applied to this graph enhance data sensemaking: (1) PageRank identifies representative data points, (2) k-core decomposition reveals dense core regions versus sparse periphery, and (3) clustering coefficient detects tight-knit neighborhoods with highly-similar data points. Through quantitative and qualitative evaluation on MNIST and Fashion MNIST, we show that these graph-based analyses are not only practical but also competitive with or complementary to purpose-built methods (e.g., k-medoids for exemplar selection, HDBSCAN for density-based clustering).