机器学习学术速递[8.24]

AI 总结：本文提出BF1，一种结合局部邻域、全局首块与对数间隔历史块的稀疏注意力方案，改造Qwen3-0.6B部分注意力层后，在长上下文场景下显著降低模型延迟，且语言建模性能优于基线方法。

链接：https://arxiv.org/abs/2608.20427

作者：Hina Dixit

英文摘要：Dense causal attention remains expensive at long context even when implemented with highly optimized exact kernels. We study BF1, a deterministic block-aligned dyadic sparse-attention route that combines a small exact local neighborhood, a global first block, and logarithmically spaced historical blocks. The route is related to prior log-sparse and dilated attention patterns; our contribution is a correctness-gated pretrained-model retrofit, a matched topology-control study, and a systems characterization that connects per-layer sparsity to whole-model latency. For fixed block width, every converted layer uses O(n log n) selected token interactions and has O(log n) graph communication depth. On an NVIDIA RTX PRO 6000 Blackwell GPU, an optimized BF16 implementation crosses dense attention between 2K and 4K tokens and reaches a 10.91x per-layer prefill speedup at 32K. Retrofitting eight of 28 Qwen3-0.6B attention layers lowers warm whole-model time to first token by 7.7%, 11.3%, and 15.3% at 8K, 16K, and 32K, respectively, while the remaining dense layers keep the complete model asymptotically quadratic. Under a matched 1,000-step, 16.384M-token adaptation protocol, BF1 ranks first across three training seeds: mean report perplexity is 1.68639 versus 1.69154 for a matched static-random nonlocal graph, 1.69258 for dense continued training, and 1.81505 for equal-budget local sliding. At seed 1234, the packed-report paired interval places Dense-CT 0.3169-0.4055% above BF1 and static-random graph 17 0.2441-0.3642% above BF1. These results establish BF1 as a reproducible sparse operator and selective retrofit primitive with real long-context systems value. This paper evaluates numerical correctness, selected-interaction scaling, kernel performance, partial-model inference, and matched next-token language modeling.

AI 总结：该研究针对大语言模型量化时自注意力机制对离散化误差敏感的问题，提出雅可比引导噪声注入训练策略，在低比特量化下显著提升了模型在图像分类与语言建模任务中的性能。

链接：https://arxiv.org/abs/2608.20988

机构：Amazon（亚马逊）

作者：Deepanshu Pandey, Arnav Chavan, Nahush Lele, Sankalp Dayal, Deepak Gupta

英文摘要：Quantization of Large Language Models (LLMs) is often hindered by the sensitivity of the self-attention mechanism to discretization errors. We identify the softmax operator as a bottleneck for quantization stability due to its sensitivity to outliers and state-dependent Jacobian. We theoretically establish that suppressing the norm of this Jacobian helps in bounding quantization-induced performance degradation. Based on this, we propose Jacobian-Guided Noise Injection, a training strategy that injects zero-mean Gaussian noise into pre-attention logits, with variance derived directly from the Jacobian Frobenius norm. Unlike prior approaches that rely on heuristic or penalise jacobian directly, our method provides a way to identify the optimal noise variance based on the local attention sensitivity. We evaluate the method on SOTA LLM architectures, where it demonstrates improved robustness over popular PTQ methods. Empirical analysis reveals that the proposed method gives up to +37% relative gains on Top-1 accuracy on ImageNet-1K for SigLIP and improves relative perplexity by upto 40% on WikiText for language models in low bit quantisation settings, proving the efficacy of the approach.

AI 总结：提出基于自由概率核的零展开超参数选择方法，无需大量展开即可高效选择储备池计算的超参数，性能接近穷举搜索且成本大幅降低，在多个时间预测任务中表现优异。

链接：https://arxiv.org/abs/2608.20998

机构： Telenor Research & Innovation（挪威电信研究与创新中心）； University of Pisa（比萨大学）

作者：Sara Malacarne, Andrea Ceni, Claudio Gallicchio

英文摘要：Reservoir computing (RC) couples a fixed recurrent dynamical system with a trained lightweight readout, but this efficiency is partly lost during hyperparameter selection: the recurrent gain, input scale, and leakage rate determine the reservoir's stability and temporal processing regime and are usually tuned through many rollouts. We introduce a deterministic, pilot-informed selector for leaky linear reservoirs followed by coordinate-wise nonlinear features. Free probability yields cross-lag propagation coefficients that summarize how the reservoir mixes past inputs. In the large-width limit, these coefficients define a deterministic temporal kernel that approximates the finite-reservoir feature geometry. Kernel ridge regression on a short labelled pilot sequence therefore ranks candidate operating regimes without instantiating or rolling out a reservoir, and the selected configuration transfers across widths. Across ten synthetic temporal benchmarks, zero-rollout selection obtains a mean deployment score of $0.772$, compared with $0.774$ for exhaustive simulation-based search, while avoiding $156\,600$ selection rollouts. With a small rollout budget, the proposed ranking provides the strongest mean performance at every tested budget and reaches the exhaustive reference using $4.8\%$ of its rollout cost. On four public electricity-transformer-temperature (ETT) forecasting datasets, five retained candidates recover the exhaustive operating point on three datasets. On multivariate cellular-traffic forecasting, 15 rollouts per cell reach the 462-rollout exhaustive reference and outperform random search and Bayesian optimization at low budgets. These results position free-probability kernels as deterministic surrogates for selecting reservoir operating regimes when validation rollouts are scarce.

AI 总结：本文提出Tydra混合架构，平衡了表格基础模型的性能与效率，在30个OpenML数据集上较TabPFN提速30%，且优于规模大10倍的Hydra模型，为表格基础模型提供了新方向。

链接：https://arxiv.org/abs/2608.21199

作者：Mieszko Komisarczyk, Saurabh Mathur, Maurice Kraus, Sriraam Natarajan, Kristian Kersting

英文摘要：Transformer-based tabular foundation models such as TabPFN achieve strong predictive performance but incur quadratic computational cost with context length. On the other hand, subquadratic SSM-based alternatives such as Hydra trade away accuracy for efficiency. To balance both, we introduce Tydra, a hybrid Transformer-State Space Model (SSM) architecture for tabular in-context learning that interleaves attention and SSM layers. Across 30 OpenML datasets, Tydra reduces inference time by 30% relative to TabPFN while retaining much of its predictive performance. Tydra also outperforms an approximately ten-times-larger Hydra model while providing faster inference. The results indicate that hybrid architectures are a promising direction for tabular foundation models.

AI 总结：本研究结合光谱滤波、降维方法与聚类算法，从40万余条牛奶中红外光谱数据中鉴定出与泌乳早期奶牛能量负平衡程度相关的五个元聚类，发现经典PCA结合k均值的高效方法可重现复杂方法的聚类结果。

链接：https://arxiv.org/abs/2608.20653

作者：T. Touil, E. R. Paquet

英文摘要：Clustering methods have been used to identify distinct groups of milk samples, cows, or herds. Fourier-transform infrared (FTIR) spectroscopy, particularly mid-infrared (MIR) spectroscopy, has been applied to individual cow milk samples to predict various milk traits. Applying clustering directly to MIR spectral data may reveal latent groups of cows associated with milk traits or health disorders and can help prevent these conditions or monitor at-risk animals. This study aimed to identify groups of individual dairy cows in early lactation directly from milk MIR spectra and to analyze their associations with milk traits. Using a dataset of 407,632 individual milk MIR records from 3,408 commercial farms, we combined (i) spectral filtering that selects informative wavenumbers, (ii) two dimensionality-reduction methods: principal component analysis (PCA) and an autoencoder, and (iii) two clustering algorithms: k-means and spectral clustering to yield eight different clustering approaches. We regrouped the assigned clusters into meta-clusters that encompassed the most similar ones identified by the eight approaches. Our results revealed five distinct meta-clusters of early-lactation individual dairy cows significantly associated with milk traits. Despite substantial differences, the eight approaches converged on the same five meta-clusters, and the classic, computationally efficient PCA-based k-means approach using the full spectrum recaptured clusters identified by more sophisticated, computationally intensive approaches. The five meta-clusters were strongly associated with DIM and appeared to reflect a gradient of negative energy balance (NEB) severity: severe, moderate, and possibly mild, while the remaining two likely represented cows recovering from NEB, one with rapid restoration of energy balance and one in early recovery.

AI 总结：该研究提出等相位自监督目标与无额外参数的Winder架构，在PTB-XL数据集上以约1M参数实现与SOTA自监督方法相当的诊断准确率，验证了编码心脏相位对称性的有效性。

链接：https://arxiv.org/abs/2608.21147

作者：Blaise Delaney, Dominic Dootson, Juan Jose Juan Castella, Salil Patel, Andrew Pfaff, Yuji Xing, Jonny Hancox, Karin Sevegnani

英文摘要： The cyclic structure of physiological processes offers a natural prior for self-supervised representation learning, and the cardiac cycle provides a particularly well-defined setting in which to exploit it. We derive a phase-equivariant self-supervised objective and introduce Winder, a joint-embedding architecture that organises representations into phase-invariant coordinates and phase-rotating harmonic subspaces. Its transport operator is fixed and closed-form, derived from the cycle's geometry rather than learned, and adds no parameters. Evaluated on PTB-XL under a frozen linear-probe protocol, Winder attains diagnostic accuracy within the range reported by state-of-the-art self-supervised methods at a ~1 M parameter footprint, while exhibiting phase-equivariant latent geometry. These findings demonstrate that explicitly encoding cardiac-phase symmetry can preserve diagnostically useful information while yielding a latent geometry that is legible, parameter-efficient, and directly tied to a measurable physiological quantity.

AI 总结：针对传统HVAC系统无法捕捉个体生理变异性的问题，提出整合多模态感知与强化学习的两阶段个性化热舒适方法，助力开发更响应的建筑控制策略。

链接：https://arxiv.org/abs/2608.20423

机构：Nottingham Trent University（诺丁汉特伦特大学）； York University（约克大学）

作者：Isibor Kennedy Ihianle, Emmanuel Manu, Ehsan Asnaashari, Mojgan Jadidi, Pedro Machado, Amrit Sagoo, Ahmad Lotfi

英文摘要：Personalised thermal comfort is essential for occupant wellbeing and for the development of more responsive building-control strategies, yet conventional Heating, Ventilation, and Air Conditioning (HVAC) systems rely on static setpoints and population-level comfort models that fail to capture individual physiological variability. This paper presents a two-stage personalised thermal comfort approach integrating multimodal physiological and environmental sensing with reinforcement learning-based decision-making.

AI 总结：该研究针对通用离散状态空间的连续时间跳跃马尔可夫决策过程，建立q学习理论基础并开发无模型算法，其在网络动态定价应用中优于基准方法。

链接：https://arxiv.org/abs/2608.20680

机构：The Chinese University of Hong Kong（香港中文大学）； University of Toronto（多伦多大学）

作者：Huiling Meng, Ningyuan Chen, Xuefeng Gao

英文摘要：We study reinforcement learning (RL) in Continuous-Time Jump Markov Decision Processes (CTJMDPs) featuring general discrete state spaces (which need not possess a vector space structure) and continuous/discrete action spaces. The setup covers many well-known applications in operations such as multi-product dynamic pricing with capacitated resources (Gallego and van Ryzin 1997). To model the exploration-exploitation tradeoff, we formulate an entropy-regularized continuous-time control problem with stochastic policies. Recent continuous-time RL techniques such as $q$-learning for controlled diffusions in (Jia and Zhou 2023) focus on continuous state spaces $\mathbb{R}^d$ and rely heavily on semimartingale theory in $\mathbb{R}^d$ for their theoretical analysis. Consequently, their methods cannot be directly applied to CTJMDPs with general discrete state spaces, which may lack the algebraic addition and subtraction structures inherent to Euclidean spaces. To bridge this gap, we establish the theoretical foundations of $q$-learning for CTJMDPs and develop model-free $q$-learning algorithms. Compared to naïve time discretization and approximating CTJMDPs using discrete-time MDPs, our approach has several conceptual and empirical benefits. Numerical experiments in network dynamic pricing (Gallego and van Ryzin 1997) show that our proposed RL algorithm reliably learns near-optimal policies and consistently outperforms standard benchmark methods, demonstrating superior solution quality and effective scalability to large-scale network instances.

AI 总结：该研究针对离线强化学习中演员与评论员耦合训练的问题，提出将策略改进与演员训练解耦的范式，通过演员生成动作候选、评论员选择，实验显示其性能优于行为克隆及联合训练的离线RL方法。

链接：https://arxiv.org/abs/2608.20909

机构：Simplexity Robotics（森普莱西蒂机器人公司）； Rensselaer Polytechnic Institute（伦斯勒理工学院）； Northeastern University（东北大学）

作者：Xuyao Lin, Yixiang Shan, Jinru Duan, Tao Yang, Xinyu Zhao, Runyu Lei, Yiming Zhao, Jiaxin Fan, Zongbao Feng, Peng Jia

英文摘要：Offline RL methods commonly jointly train the actor and critic, where the critic is used to guide the actor toward higher-value actions. This coupled learning process is well motivated in online RL, where an improved actor collects new data that can further update the actor and the critic. However, training data remains fixed in offline RL, making actor-side policy improvement unable to generate new data to validate or correct the critic. Moreover, retaining this coupled paradigm leads to two related challenges. Firstly, actor updates can drift toward high-valued but potentially out-of-distribution (OOD) actions and amplify critic overestimation. Secondly, conservative value estimation or behavior-cloning regularization creates a difficult trade-off between suppressing OOD actions and selecting high-value actions within the data-supported region. Motivated by this observation, we revisit the conventional offline RL paradigm and propose decoupling policy improvement from actor training. Specifically, we train the actor solely to model the behavior distribution and perform policy improvement at inference time by reranking multiple actor-generated proposals with a separately learned critic. We refer to this paradigm as the decoupled policy extraction paradigm. Under such paradigm, the actor provides behavior-supported action candidates, while the critic performs value-based selection within this candidate set. Extensive experiments show that the decoupled policy extraction paradigm outperforms both behavior cloning and jointly learned offline RL methods, while remaining effective even with a naive Q-learning critic.

AI 总结：TracingFlow是一种无模拟的二阶动力学轨迹推断框架，通过回归加速度场解决动力最优加速度传输问题，在合成与scRNA-seq数据集上实现了更优的分布重构与轨迹保真度，还能恢复兼具数学最优性与生物学合理性的动力学结构。

链接：https://arxiv.org/abs/2608.21070

机构：Center for Machine Learning Research, Peking University（北京大学机器学习研究中心）； School of Mathematical Sciences, Peking University（北京大学数学科学学院）； Center for Quantitative Biology, Peking University（北京大学定量生物学中心）； National Engineering Laboratory for Big Data Analysis and Applications, Beijing（北京大数据分析与应用国家工程实验室）； AI for Science Institute, Beijing（北京科学智能研究院）

作者：Yuhao Sun, Zekun Wu, Zixun Huang, Peijie Zhou

英文摘要：Inferring continuous system evolution from sparse temporal snapshots is a key challenge in generative modeling and single-cell omics. While Optimal Transport (OT) is popular, existing frameworks are largely restricted to first-order dynamics, assuming memoryless velocity fields. This limits expressiveness, as first-order systems fail to account for regulatory momentum and time-delayed responses inherent in processes like cell differentiation. Here, we introduce TracingFlow, a simulation-free Flow Matching framework generalizing to second-order dynamics. By using neural networks to regress the acceleration field, TracingFlow provides an exact, efficient solution to the Dynamical Optimal Acceleration Transport (DOAT) problem. Unlike first-order methods yielding over-smoothed trajectories, our second-order formulation captures high-curvature transitions and nonlinear evolutions by learning the underlying force fields. Evaluated on complex synthetic and large-scale scRNA-seq datasets, TracingFlow achieves superior accuracy in distributional reconstruction and trajectory faithfulness. Moreover, by integrating lineage tracing priors, it recovers dynamical structures that are both mathematically optimal and biologically plausible.

AI 总结：本文提出神经符号框架，将LLM作为自适应提议分布，结合其先验知识与数据经验评分，在真实临床数据集上建模APOs，恢复专家验证边并识别新因果关系，为干预提供新见解。

链接：https://arxiv.org/abs/2608.21079

机构：The University of Texas at Dallas（德克萨斯大学达拉斯分校）； Technical University of Darmstadt（达姆施塔特工业大学）； Swarthmore College（斯沃斯莫尔学院）； Indiana University School of Medicine（印第安纳大学医学院）； Hessian Center for Artificial Intelligence（黑森人工智能中心）； German Research Center for AI (DFKI)（德国人工智能研究中心（DFKI））

作者：Kavimayil P. Komarasamy, Saurabh Mathur, Ameet Soni, David M. Haas, Kristian Kersting, Sriraam Natarajan

英文摘要：Adverse Pregnancy Outcomes (APOs) such as preterm birth and gestational diabetes can have long-term consequences for both the mother and child, yet an understanding of their causes remains elusive. Causal discovery in this domain is especially challenging due to a paucity of data and incomplete domain knowledge. As a result, pure data-driven methods fail, and Large Language Model (LLM) outputs remain inconsistent or contradictory. We introduce a neurosymbolic framework for generating plausible causal hypotheses that iteratively combines the broad prior knowledge of LLMs with empirical scoring on data. Our method treats the LLM as an adaptive proposal distribution, generating hypotheses that are scored against empirical data; the resulting high-scoring graphs are then used to update the LLM's context, steering subsequent generations toward more promising regions of the hypothesis space. We evaluate our approach on a real-world clinical dataset for modeling APOs and their risk factors, comparing our results against an expert-constructed causal graph. Our method recovers all expert-validated edges and identifies additional plausible causal relations not previously listed by experts, potentially providing new insights for targeted interventions.

AI 总结：本研究提出整合多技术的破产预测框架，在不平衡金融数据中优化少数类检测，经实验验证了模型性能，可助力财务困境企业的早期预警系统建设。

链接：https://arxiv.org/abs/2608.20343

作者：Obu-Amoah Ampomah, Edmund Fosu Agyemang, Kofi Acheampong, Louis Agyekum, Enock Adu Bonsu, Eric Nyarko

英文摘要： This study develops and evaluates a bankruptcy prediction framework that integrates consensus-based feature selection, hybrid resampling, stacking ensembles, and explainable artificial intelligence to improve minority-class detection in severely imbalanced financial data. Using the Taiwanese Bankruptcy Prediction dataset from the UCI Machine Learning Repository, five feature-selection algorithms were first applied, and a consensus retention rule reduced the input space to 23 robust variables. The balanced training data were then generated using SVM-SMOTE, SMOTE-Tomek, and SMOTE-ENN. Five ensemble machine learning classifiers, namely gradient boosting, extreme gradient boosting, histogram-based gradient boosting, LightGBM, and AdaBoost, were compared with five deep learning models, including RNN, LSTM, GRU, DNN, and MLP. In addition, hybrid stacking ensembles combined the five machine learning classifiers as base learners with each deep learning model as a meta-learner. Model performance was assessed using accuracy, recall, specificity, G-mean, and ROC-AUC, while SHAP was used to explain feature contributions. The results show that resampling strategy materially shaped model behavior. SVM-SMOTE and SMOTE-Tomek favored accuracy and specificity, whereas SMOTE-ENN delivered stronger minority-class detection. Among standalone models, the GRU with SMOTE-ENN achieved the best overall predictive balance, with recall of 0.8627, G-mean of 0.8517, and ROC-AUC of 0.9431. Among stacking ensembles, SMOTE-ENN with (GB+XGB+HGB+LGBM+AB)+LSTM provided the strongest compromise between sensitivity and specificity. SHAP analysis identified leverage, profitability, solvency, and operational efficiency indicators as the most influential predictors of bankruptcy risk. These findings support more reliable and interpretable early warning systems for financially distressed firms.

AI 总结：该研究揭示阈下特质迁移的两阶段机制，优化器状态传输源扰动，后续训练决定其行为价值，通过Qwen、Llama等模型实验验证了该因果机制。

链接：https://arxiv.org/abs/2608.20442

机构：Xiamen University（厦门大学）

作者：Qinyang Xu

英文摘要：Subliminal trait transfer allows a student model to acquire behavioral dispositions from teacher-generated data in which the trait is not semantically expressed. Recent work explains how such signals enter gradients, but not how they survive source removal or acquire different signs under later training. We treat parameters and optimizer moments as a single trainer state and derive an exact transport-valuation identity separating observer-independent propagation of the source perturbation from the value assigned by a future continuation and behavioral readout. State surgery identifies the first moment as a causal carrier. Transplanting it alone leaves parameters, hidden states, and outputs unchanged at the cut, yet source-free updates generate growing parameter and hidden-state differences; transplanting parameters with the first moment recovers the terminal behavioral response. Sending the same source-induced difference through matched futures produces negative, near-zero, and positive Qwen effects (-0.658, +0.008, and +0.658 seed means). This ordering recurs in all 12 Llama-3.2-1B seeds after eight updates, while state-difference norms remain nearly equal across routes. Both contrasts grow in every paired seed when the continuation extends to sixteen updates. A full-horizon costate predicts all 42 Qwen route-mean signs and all 21 resolved Llama ordinary-route signs. Observer-independent transport also replicates across Qwen, SmolLM2, and Llama, while the complete-state recurrence predicts physical, hidden, and fixed-head responses in non-LoRA MNIST systems, including CNNs trained with AdamW and momentum SGD. Together, these results identify a two-stage mechanism for subliminal trait transfer: optimizer state transports the source perturbation, and later training determines its behavioral value.

AI 总结：本研究通过银行电话营销案例对比了互信息与基于数据的敏感性分析两种特征选择方法的优劣，发现二者在不同假阳性场景下各有优势，互信息仍为有效方法且敏感性分析可用更少特征取得良好预测结果。

链接：https://arxiv.org/abs/2608.20447

机构：Universidad Nacional de Tres de Febrero（阿根廷 Tres de Febrero 国立大学）； Instituto Universitário de Lisboa (ISCTE-IUL)（里斯本大学学院）； ISTAR-IUL（ISTAR-IUL 研究所）； ALGORITMI Research Centre（ALGORITMI 研究中心）； University of Minho（米尼奥大学）； Dataxplore； Boldt Gaming

作者：Nestor Barraza, Sergio Moro, Marcelo Ferreyra, Adolfo de la Peña

英文摘要：Feature selection is a highly relevant task in a data-driven knowledge discovery project. Several techniques have been developed aiming at finding the features that influence most an outcome to predict, including mutual information and, in recent years, the data-based sensitivity analysis. The present research focus on analyzing the advantages and disadvantages of each of these two techniques, by applying both to a bank telemarketing case. Thereafter, a logistic regression model is built on the tuned set of features identified by each of the two techniques as the most influencing set of features on the success of a telemarketing contact, in a total of 13 features for mutual information and 9 features for the data-based sensitivity analysis. The latter performs better for lower values of false positives while the former is slightly better for a higher false positive ratio. Thus, mutual information becomes a better choice if bank managers intend to reduce slightly the cost of contacts without risking losing a high number of successes. Such results show that mutual information, although not recent, is still a valid method for feature selection. On the other side, the data-based sensitivity analysis selection achieved good prediction results with less features.

AI 总结：本文提出Bern2Edge端到端框架，通过知识蒸馏将预训练教师网络转为硬件高效的伯恩斯坦多项式激活表示，支持两种部署路径，在边缘FPGA上实现了延迟、资源占用降低，同时保持较高准确率。

链接：https://arxiv.org/abs/2608.20497

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

作者：Malak Gamal El-Din, Yifan Zhang, Yasser Shoukry, Sitao Huang, Salma Elmalaki

英文摘要： Deploying high-accuracy neural networks on resource-constrained edge devices remains challenging, as existing approaches treat training, compression, and hardware synthesis as separate stages, leaving a gap between software-trained models and efficient end-to-end deployment with limited support for interpretability. We propose Bern2Edge, an end-to-end framework that uses knowledge distillation to convert a pretrained teacher feed-forward network into hardware-efficient representations via Bernstein polynomial activations. This representation enables two deployment paths: (i) a high-fidelity LUT-based realization that preserves model fidelity under compression, and (ii) a symbolic rule-based representation derived from Bernstein activation geometry, enabling interpretable inference with explicit input-space constraints. The resulting BNNs achieve up to 2.12 percentage-point (pp) accuracy improvement over ReLU under identical compression constraints. At the system level, Bern2Edge achieves up to 99.8% latency reduction and 95.2% BRAM reduction relative to a W8A8 quantized teacher on an AMD Xilinx KV260 FPGA, while maintaining accuracy within 0.5 pp, and further deploys on a low-power Spartan-7 XC7S15 FPGA. The rule-based path reduces DSP usage by up to 89.0% at a cost of 1.5 pp in total accuracy.

AI 总结：该研究针对图联邦学习中客户端图结构异质性问题，提出FlatLand方法，通过定制洛伦兹空间与参数解耦策略实现个性化联邦学习，在低维设置下表现更优。

链接：https://arxiv.org/abs/2608.21096

机构：The Chinese University of Hong Kong（香港中文大学）； Yale University（耶鲁大学）； Huawei Technologies Co., Ltd.（华为技术有限公司）； The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））

作者：Jiahong Liu, Ram Samarth B B, Xinyu Fu, Menglin Yang, Weixi Zhang, Rex Ying, Irwin King

英文摘要：Federated learning enables privacy-preserving collaborative training, but highly heterogeneous client data remain challenging, especially in graph federated learning where clients possess structurally diverse graphs. Existing personalized federated learning (PFL) methods ignore the intrinsic geometric properties of diverse graph structures. We propose FlatLand, a novel personalized federated learning method that embeds different clients' data in tailored Lorentz space of hyperbolic geometry. Our key insight is that hyperbolic geometry naturally accommodates the intrinsic negative curvature prevalent in real-world graphs, while the time-like dimension in Lorentz space provides a principled way to encode client-specific heterogeneity. We develop a parameter decoupling strategy that separates heterogeneous information (captured in time-like parameters) from common knowledge (preserved in space-like parameters), enabling direct aggregation without requiring client similarity estimation and extra calculation modules. Empirical results on diverse federated graph learning tasks demonstrate that FlatLand achieves superior performance, particularly in low-dimensional settings.

AI 总结：BackDFL是评估去中心化联邦学习（DFL）在自适应后门攻击下安全性的统一基准，实验表明现有鲁棒方法在15%恶意参与率下即失效，且性能随通信拓扑变化大。

链接：https://arxiv.org/abs/2608.21137

机构：SAMOVAR, Télécom SudParis, Institut Polytechnique de Paris（萨莫瓦尔实验室、南巴黎电信学院、巴黎综合理工学院）； PIRAT, INRIA Rennes（皮拉特实验室、法国国家信息与自动化研究所雷恩分部）

作者：Mouhamed Amine Bouchiha, Gregory Blanc, Yufei Han

英文摘要：Decentralized Federated Learning (DFL) promises trust-free collaborative learning by replacing the centralized parameter server with peer-to-peer model exchange. However, this architectural shift fundamentally reshapes the threat landscape. Without globally coordinated aggregation, DFL becomes particularly susceptible to backdoor attacks, in which malicious participants implant persistent hidden behaviors while maintaining high clean-task performance. In this paper, we argue that the robustness of DFL has been significantly overestimated. Existing studies rely on simplified threat models, non-adaptive adversaries, fragmented evaluation protocols, inconsistent communication topologies, and ad hoc training configurations, leading to an incomplete understanding of DFL security. To address these limitations, we present BackDFL, a unified benchmark for systematically evaluating DFL under realistic and adaptive backdoor attacks. Through extensive experiments, BackDFL exposes critical failure modes of decentralized learning. Our results demonstrate that both state-of-the-art Byzantine-robust DFL methods and adapted FL backdoor defenses fail under modest malicious participation rates (as low as 15%), especially in heterogeneous settings, while their robustness varies substantially across communication graph topologies.

AI 总结：该研究针对CNN的对抗脆弱性与高训练成本问题，提出在训练时对GPU降电压，利用硬件故障作为隐式正则化，在不修改算法的情况下提升对抗鲁棒性并降低能耗。

链接：https://arxiv.org/abs/2608.20572

机构：George Mason University（乔治梅森大学）； SecureMind Technologies Inc（SecureMind科技公司）； Rochester Institute of Technology（罗切斯特理工学院）

作者：Behnam Omidi, Ahmad Tahmasivand, Husam Alsyouri, Saba Al-Sayouri, Chongzhou Fang, Ihsen Alouani, Khaled N. Khasawneh

英文摘要： Convolutional Neural Networks (CNNs) face a dual challenge: vulnerability to adversarial attacks and prohibitive training cost. Adversarial training is effective but expensive, a burden that grows as learning shifts to the energy-constrained edge. This paper addresses both through GPU undervolting during training. Reducing supply voltage introduces stochastic perturbations that act as implicit regularization, improving robustness while lowering power. We characterize undervolting-induced faults at the bit level, then train LeNet, VGG-6, and MobileNetV3 on MNIST and CIFAR-10 under two training regimes, standard and adversarial, each at nominal and undervolted voltage, and evaluate all models against adversarial attacks. In both regimes, the undervolted model consistently achieves higher adversarial accuracy than its nominal-voltage counterpart, showing that hardware-induced faults strengthen even adversarial training. Because dynamic power scales quadratically with supply voltage, these robustness gains arrive with substantial energy savings. GPU undervolting is therefore a readily deployable hardware-level defense requiring no algorithmic change, and opens a promising direction in which robustness and energy efficiency move together.

AI 总结：该研究针对多变量交通流预测，推出保留三类原始测量的PEMSB-3V基准，提出模型无关的RiskTraf方法，通过风险外推残差学习缓解状态依赖捷径，提升预测性能且优于相关适配方法。

链接：https://arxiv.org/abs/2608.20656

机构：Dongbei University of Finance & Economics（东北财经大学）； Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））

作者：Guangyu Wang, Zhidan Liu

英文摘要：Traffic sensors commonly record flow, speed, and occupancy, but standard traffic flow forecasting benchmarks and models rarely exploit all three raw measurements reliably. Although speed and occupancy provide sensor-native traffic-state information beyond flow alone, existing releases often omit these variables, replace them with proxies, or contain logically inconsistent records. Moreover, direct empirical risk minimization over three-variable inputs may exploit regime-dependent shortcuts, as the relationships among flow, speed, and occupancy vary substantially between free-flow and congested states. We introduce \textbf{PEMSB-3V}, a public benchmark suite that preserves raw flow, speed, and occupancy measurements from PeMS detectors for flow prediction. We also propose \textbf{RiskTraf}, a model-agnostic risk-extrapolated residual plug-in. For each trained spatio-temporal backbone, RiskTraf freezes the selected checkpoint and learns a lightweight zero-start residual head from historical speed and occupancy. The residual head constructs ordered traffic-risk environments and optimizes horizon-wise flow corrections with a risk extrapolation objective, thereby mitigating regime-specific shortcut correlations without modifying the backbone. Extensive experiments demonstrate that RiskTraf consistently improves diverse forecasting backbones and outperforms debiasing and distribution-shift adaptation methods. Our code and benchmark are available at this https URL.

AI 总结：该研究针对带贝叶斯输出层的图神经网络，发现潜在-后验对齐现象，提出对齐引导学习方法，可降低预测不确定性并改善结构校准，为不确定性动力学提供新视角。

链接：https://arxiv.org/abs/2608.20758

机构：Korea Institute of Science and Technology（韩国科学技术研究院）； Clean Energy Research Center（清洁能源研究中心）

作者：Suk Hoon Choi, Damdae Park, Junhyuk Choi, Hyein Jung, Changsoo Kim, Ung Lee, Kyeongsu Kim

英文摘要：Bayesian Neural Networks (BNNs) with Bayesian output layers provide a principled and tractable framework for quantifying predictive uncertainty, yet the mechanisms shaping that uncertainty remain unclear. While conventional theory attributes uncertainty reduction to posterior contraction, the corresponding assumptions need not hold for deep models. In the Graph Neural Networks (GNNs) with Bayesian output layers studied here, we observe that predictive uncertainty decreases as latent representations shift toward lower-variance posterior directions, even though the posterior variance does not contract. We term this behavior Latent-Posterior Alignment (LPA) and conduct interventional experiments that support its functional role in shaping predictive uncertainty. Building on this insight, we propose Alignment-Guided Learning (AGL), which explicitly promotes this alignment during training. AGL effectively reduces predictive uncertainty while preserving accuracy and improves structural calibration, ensuring that the model confidence faithfully mirrors underlying data density. These findings provide a new perspective on uncertainty dynamics in GNNs with mean-field Bayesian output layers, shifting the focus from the magnitude of the posterior to the geometric interplay between latent and parameter spaces.

AI 总结：本文审查了时空预测领域的基准数据集与基线，发现空间无关线性模型表现优于预期，揭示了一阶差分数据集的结构偏差，建议减少对这类数据集的依赖并提出了应用分析结果开发GNN模型的新途径。

链接：https://arxiv.org/abs/2608.20980

机构：Imperial College London（帝国理工学院）； Ruhr University Bochum（鲁尔大学波鸿分校）； Queen Mary University of London（伦敦玛丽女王大学）； Ben-Gurion University of the Negev（内盖夫本-古里安大学）

作者：Kenneth Martin, Simon Heilig, Asja Fischer, Michel F. C. Haddad, Adam M. Sykulski, Moshe Eliasof

英文摘要：Graph neural networks (GNNs) are routinely employed for short-range forecasting on multivariate time series with a spatial graph structure. Despite the availability of many alternative datasets, method innovations within this domain are predominantly assessed against a rather limited set of benchmark datasets, most notably Chickenpox, PedalMe, WikiMaths, METR-LA, and PEMS-BAY. The evaluation protocols contain baselines spanning from historical averages to classical machine learning approaches. These baselines often show competitive performance compared to GNNs. In the present work, we take a step back and analyse the benchmark datasets via classical time series methods to uncover why spatially-unaware linear models pose a stronger competitor than previously reported, casting further doubt on the discriminative reliability of the aforementioned widely adopted datasets. Our statistical analysis provides a toolset for identifying significant spatial and temporal correlations, while revealing a structural bias introduced by first-order differenced datasets. We therefore recommend reducing the over-reliance on such datasets for method comparison, and instead advocate for more rigorous statistical evaluation. By applying the results of our analysis to a simple hybrid model, we show how our methodology can lead to novel ways of developing GNN models

AI 总结：Thermo-FL是热感知联邦LoRA微调框架，通过客户端温度调控和服务器TERRA鲁棒聚合，提升边缘LLM在热应力与攻击下的鲁棒性，稳定温度并保留任务性能。

链接：https://arxiv.org/abs/2608.21172

机构：Kennesaw State University（肯尼索州立大学）

作者：Shiva Shrestha, Kazi Shaharair Sharif, Zongxing Xie, Jiajing Huang, Anhao Xiang, Honghui Xu

英文摘要：Federated fine-tuning enables large language models to adapt on edge devices without centralizing private data, but practical deployments must address hardware instability and adversarial update corruption together. Thermally constrained clients may throttle, slow local training, or delay synchronous aggregation, while Byzantine clients and communication-layer adversaries can corrupt the updates used to form the global model. To address these challenges, we present Thermo-FL, a thermal-aware federated LoRA fine-tuning framework that uses device temperature as an active control signal for local adapter training and sparse update transmission. On the client side, Thermo-FL adjusts the active LoRA-layer fraction and transmitted update density as devices heat or cool, reducing workload under thermal stress. On the server side, Thermo-FL introduces TERRA, a robust aggregation pipeline for dynamically sparse LoRA updates that combines norm filtering, mask-aware directional validation, adaptive active-coordinate clipping, and mask-aware aggregation. We evaluate Thermo-FL using both a large-scale emulator and a Jetson-based physical testbed. In the emulator, Thermo-FL improves robustness under adversarial sparse aggregation and achieves the strongest BoolQ accuracy across clean and attack settings while remaining competitive on GSM8K. In the physical prototype, Thermo-FL stabilizes device temperature, reduces compressed upload size through bitmap sparse encoding, and preserves GSM8K utility under sign-flip/scale and MITM perturbations. These results show that secure edge LLM adaptation should jointly consider hardware behavior, workload regulation, sparse communication, and aggregation robustness.

AI 总结：SPARCL 是一种谱分区解析式持续学习方法，通过分解自相关为高能核心与残差，冻结旧类别分类器核心，更新残差块，在多数据集上缩小了经典解析式学习器与强表示匹配器的性能差距，且与 Fly-CL 等方法互补。

链接：https://arxiv.org/abs/2608.21307

机构：University of Sheffield（谢菲尔德大学）

作者：James Hartley, Zeropy Surio, Daniel Whitmore, Hannah Clarke, Thomas Reed

英文摘要：Analytic continual learning has emerged as a strong exemplar-free alternative to gradient-based class-incremental learning because it replaces iterative optimization with closed-form ridge updates. Yet the usual forgetting narrative, centered on stochastic gradient overwriting, does not explain why analytic methods still drift on old classes despite exact recursive solvers. We identify the culprit as spectral interference: the joint ridge classifier for all tasks shares the inverse autocorrelation operator $(R+\lambda I)^{-1}$, so incoming task samples that load onto old dominant eigendirections dilute the spectrum and perturb old-class logits even when old labels are never revisited. Based on this view, we propose SPARCL, a spectral partitioned analytic continual learner that decomposes the running autocorrelation into a high-energy core and a residual complement, freezes old-class classifier components in the core subspace, and updates only the residual block through recursive least squares with an optional residual random-projection expansion. This yields a simple closed-form update with a provable invariance guarantee for the core contribution of old logits. Across CIFAR-100, CUB-200, ImageNet-R, and ImageNet-A under a frozen ViT-B/16 protocol, SPARCL closes most of the gap from classical analytic learners to strong representation matchers, while remaining complementary to sparse feature-decorrelation approaches such as Fly-CL.

AI 总结：本文推出Metag数据集，用于构建元评审智能体，助力元评审员快速确认作者对评审意见的回应及变更位置，提升同行评审透明度与可追溯性。

链接：https://arxiv.org/abs/2608.20488

机构：Georgia Institute of Technology（佐治亚理工学院）； Microsoft（微软公司）

作者：Anirudh Sundar, Min Chen, Divya Tadimeti, Gemma Zhang, Alice Li, Nigel Boachie Kumankumah, Pavan Uttej Ravva, Sadid Hasan, Somya Chatterjee, Pruthvi Prakash Navada, Xiao Wang, Yue Kang, Sulaiman Vesal, Larry Heck

英文摘要： AI tools increasingly support tasks across the scientific research cycle, from experiment design and manuscript preparation to peer review. At the same time, the continuing growth in conference submissions has increased the burden on meta-reviewers, who must synthesize reviewer feedback, author rebuttals, and manuscript revisions. To address this concern, this paper introduces Metag, a dataset to accelerate the development of meta-reviewing agents, specifically to identify changes made to scientific articles during the review-rebuttal process. Each instance contains a reviewer concern, the author's proposed resolution, and the manuscript diffs implementing the stated change. Metag is collected by obtaining manuscript versions from before the review deadline and after acceptance, computing differences between the two documents, and asking human annotators to align these differences with action items from OpenReview discussions. The resulting dataset consists of 349 high-quality action items tied to paper differences and will enable building methods to empower meta reviewers to quickly identify whether authors have addressed reviewer statements and where in the paper those changes have been made, resulting in additional transparency and traceability throughout peer review. The dataset is publicly available at this https URL.

AI 总结：本研究评估ARIMA、随机森林、XGBoost模型，提出MLAMA集成方法，基于安大略省COVID-19数据验证其预测性能更优，支持按操作条件选择预测模型。

链接：https://arxiv.org/abs/2608.20406

机构：Public Health Ontario（安大略省公共卫生局）； Dalla Lana School of Public Health, University of Toronto（多伦多大学达拉·拉纳公共卫生学院）； University of Toronto（多伦多大学）； York University（约克大学）

作者：Yushu Zou, Ye Li, Johra Moosa, Martin Grunnill, Samir N. Patel, Venkata R. Duvvuri

英文摘要：Public health forecasts must respond to abrupt changes in surveillance data without over-extrapolating noise, reporting artifacts, or temporary trends. We evaluated autoregressive integrated moving average (ARIMA), random forest, and extreme gradient boosting (XGBoost) models using 190 weekly observations of publicly available Ontario COVID-19 case counts from January 2020 to October 2023. Rolling-origin time-series cross-validation preserved temporal order during model tuning and evaluation. Performance was assessed across three operating dimensions: responsiveness following selected turning points, forecast horizons of one to six weeks, and the amount of historical training data. We also developed Machine Learning and ARIMA Model Averaging (MLAMA), a non-negative performance-weighted ensemble with weights that vary by forecast horizon and responsiveness setting. Retrospective comparisons showed that ARIMA adapted rapidly after turning points but its normalized error increased at longer horizons. Random forest and XGBoost were less responsive initially but maintained more stable normalized error over longer horizons. For two-week forecasts at the end of the study period, training on the most recent data outperformed using longer historical periods, particularly for XGBoost. MLAMA achieved the lowest normalized mean absolute percentage error across most forecast horizons and ranked among the best-performing methods across responsiveness settings. These findings support selecting forecasting models according to operating conditions rather than relying on a single universally preferred approach. MLAMA provides a practical framework for combining complementary statistical and machine-learning forecasts. The accompanying Python package is currently maintained in a private repository while software validation and reproducibility testing are completed.

AI 总结：该研究提出跨参数重链接数据投毒原语，在神经PDE算子中构造错误物理后门，在保持低干净误差的同时实现高攻击成功率，暴露了其结构性验证缺口。

链接：https://arxiv.org/abs/2608.20439

作者：Hanbing Liang, Fujun Liu

英文摘要：Neural PDE operators are increasingly trained on reusable solver archives, yet validation often relies on clean prediction error and parameter-agnostic plausibility checks. We introduce cross-parameter relinking, a data-poisoning primitive that makes a triggered input select a valid solution from the same PDE family under an incorrect physical parameter. We term this a wrong-physics backdoor: the output remains physically plausible but is wrong for the intended parameter. The attack exploits tensor-to-parameter provenance failures in multi-parameter archives by stamping the surrogate input and relinking its supervision to a cached alternate-parameter solution for the same latent sample. Across 476 attack campaigns, we evaluate Burgers, advection-diffusion, two-dimensional Navier-Stokes, and an elliptic Poisson case. Fourier Neural Operators and DeepONet provide the primary evidence, with Transformer, GRU, and LSTM models as support. FNO reaches a backdoor success rate of 1.0000 on both advection-diffusion and two-dimensional Navier-Stokes while retaining low clean relative L2 error. Clean-label, label-only, and shuffled controls show that high attack success alone is insufficient: successful attacks must move predictions toward the intended alternate-physics target while preserving bounded clean error. These results expose a structural validation gap: smoothness or generic solver-like behavior is insufficient unless the provenance of the intended physical parameter is also verified.

AI 总结：该研究针对无高保真参考解时神经算子部署的最优选择难题，提出利用共享物理响应的方法，恢复模型库隐藏排名，准确率超99%，实现无需真值数据的高效科学代理部署。

链接：https://arxiv.org/abs/2608.20441

机构：Changchun University of Science and Technology（长春理工大学）； School of Physics（物理学院）

作者：Hanbing Liang, Fujun Liu

英文摘要：Selecting the optimal neural-operator prediction during deployment is challenging when high-fidelity reference solutions are unavailable. We demonstrate that under a squared Hilbert-space loss, ranking a finite model library depends strictly on the low-dimensional span of candidate differences, allowing us to score all models simultaneously using a single anchor-based linearized response of the governing equation. This shared physical diagnostic accurately recovered over 99.6\% of pairwise preferences and 99.0\% of optimal checkpoints across diverse Fourier and convolutional operator libraries for fluid, reaction-diffusion, and wave dynamics. Furthermore, the corrected physical proxy frequently outperformed the best individual candidates, and we establish computable sufficient conditions that rigorously certify exact decisions for strongly monotone discretizations. By exploiting the local dynamical response rather than raw defect magnitude, this framework enables the reliable and highly efficient deployment of scientific surrogates without requiring ground-truth data.

AI 总结：本文提出Fuzzy-MoE模型，通过双视图模糊路由器实现可解释专家路由，在非平稳多元时间序列预测中显著优于主流方法，同时具备透明可追溯的机制。

链接：https://arxiv.org/abs/2608.20761

作者：Lan Guo, Jie Xiao, Zhao Su, Jun Shen, Haoran Li, Weixia Ma, Qingguo Zhou, Binbin Yong

英文摘要：In non-stationary multivariate time series, different variables and samples often exhibit heterogeneous latent dynamic states, while existing deep forecasting models usually compress them into a unified end-to-end mapping, leading to suboptimal modeling of time-varying dynamics and limited interpretability regarding which forecasting mechanism is activated under different latent states. To overcome these limitations, we reformulate time series forecasting as a unified framework of latent temporal state identification and interpretable expert routing, and propose Fuzzy-MoE, a fuzzy logic-based dynamic Mixture-of-Experts model. Fuzzy-MoE consists of multiple parallel expert mapping networks and a dual-view fuzzy router. By jointly exploiting local convolutional dynamics and global segmented statistics, the router infers latent temporal states and computes expert activation strengths through learnable Gaussian membership functions, enabling explicit IF-THEN rule-based expert selection. This fine-grained routing strategy allows different variables within the same sequence to activate different experts, effectively capturing heterogeneous temporal dynamics while improving model interpretability. Experimental results on multiple public time series benchmark datasets show that Fuzzy-MoE significantly outperforms mainstream forecasting methods in forecasting accuracy. Moreover, fuzzy memberships and rule activations provide interpretable routing diagnostics, demonstrating the effectiveness of the proposed framework in both forecasting performance and mechanism transparency. Unlike traditional MoE models that use black-box routing, Fuzzy-MoE`s routing is based on clear, interpretable fuzzy rules. This makes the expert selection transparent and traceable.

AI 总结：本研究针对药物发现智能体AI，提出人类对齐的LLM作为评判者评估框架，优化后对齐度达0.86，为科学领域智能体系统评估提供可复用模板。

链接：https://arxiv.org/abs/2608.21057

机构：AstraZeneca（阿斯利康）； Chalmers University of Technology（查尔姆斯理工大学）； University of Gothenburg（哥德堡大学）； Science for Life Laboratory (SciLifeLab)（生命科学实验室）

作者：Emma Granqvist, Rocío Mercado, Samuel Genheden

英文摘要：Agentic large language model (LLM) systems are reshaping scientific workflows in chemistry and drug discovery, but evaluating their open-ended, tool-augmented outputs remains a fundamental bottleneck. Reference-based metrics such as BLEU and ROUGE fail to capture semantic correctness, while expert human evaluation does not scale to the iteration speed these systems demand. The LLM-as-a-Judge paradigm has emerged as a scalable alternative, but existing drug discovery benchmarks deploy LLM judges without validating their alignment with human experts. In this work, we present an LLM-as-a-Judge evaluation framework for ChatInvent, an agentic drug discovery assistant deployed at AstraZeneca, with four contributions. First, we define four output-quality evaluation dimensions---Completeness, Relevancy, Structural Clarity, and Scope Adherence---alongside deterministic Tool Call Correctness checks. Second, we validate the judge through a human alignment study with five expert annotators, comparing Gemini 3.1 Pro, Claude Opus 4.7, GPT-5, and Llama 3.1 70B as candidate judges. Third, we optimize the best-performing judge using few-shot demonstrations of human-annotated examples, improving alignment with the human majority vote from 0.80 to 0.86. Fourth, applying the optimized judge to 70 held-out questions, we surface concrete limitations and find that informal phrasings do not systematically degrade output quality; if anything, it is helpful to have the LLM rewrite the original question before querying the agent. Our framework provides a reusable template for human-aligned evaluation of agentic systems in scientific domains.

AI 总结：该研究针对多流运行遥测异常检测问题，提出TRACE-C秩校准检测器，经英国电网数据验证，可有效识别风暴等异常事件，同时明确了方法的解释限制。

链接：https://arxiv.org/abs/2608.21251

作者：Matthew Faucher

英文摘要：Operational telemetry can be jointly anomalous while every individual stream stays inside its familiar range. TRACE-C is an auditable strictly-prior rank-calibrated detector for aligned multi-stream telemetry: same-regime rolling median/MAD residuals feed three window channels -- a maximum normalized local sum, a Gaussian copula-form dependence contrast on robust-z residuals, and a worst standardized AR(1) innovation -- whose channel ranks are Fisher-aggregated and ranked against earlier aggregates. We evaluate six Great Britain grid streams with a January-April 2019 fit, July-December 2019 development evidence, and a 2020 hold-out frozen before inspection. TRACE-C ranks Storm Atiyah first among 2019 test windows, but a disclosed channel ablation attributes that rank to the local channel, not the copula-form channel: copula-only ranks Atiyah 59th. The short 9 August frequency event is ranked far lower by the fused detector (143) than by the temporal channel alone (40), and reconstruction baselines rank it first. In 2020 no window is selected, which is consistent with record-rule saturation rather than an uneventful year; the highest-ranked frozen window was later interpreted as Storm Ellen. Three interpretive limits carry throughout. The resulting p-values are selection quantities, not event probabilities. The copula-form channel is not a literal copula density: the method applies no probability-integral or normal-score transform. Empirical rank counts are diagnostics, not coverage or false-discovery proofs. Every table and figure in this paper is generated from committed machine-readable reports.

AI 总结：ConceptTS是围绕人类可理解概念的可解释多变量时间序列预测框架，利用大语言模型生成概念与标注规则，在空气质量数据集上达到与黑箱基线相当的准确率，且决策过程可解释。

链接：https://arxiv.org/abs/2608.21277

机构：Stanford University（斯坦福大学）； University of California, Davis（加州大学戴维斯分校）

作者：Yichen Jiang, Yueqiao Chen, Dongyu Liu

英文摘要：State-of-the-art multivariate time-series forecasters can model complex temporal and cross-variable dependencies, yet their opaque representations provide limited insight into why a particular forecast is produced. This lack of transparency restricts their use in settings where practitioners must understand and assess the factors underlying a prediction. We introduce ConceptTS, an interpretable forecasting framework that organizes its predictions around named, human-readable concepts. ConceptTS uses a large language model to propose task-relevant concepts and generate executable labeling rules, translating the language model's domain knowledge into direct supervision without costly manual concept annotation. The proposed concepts are organized into three complementary bottlenecks that describe the historical context, local forecast intervals, and the full forecast horizon. A shared decoder combines representations derived from their predicted activations to construct the forecast, making the model's decision process explicit and supporting direct concept-level interventions. Experiments on the Beijing Multi-Site Air Quality dataset show that ConceptTS achieves accuracy competitive with strong black-box baselines while producing semantically meaningful concept activations.

AI 总结：本文研究受控随机过程（含变换器）极小表示的扰动稳定性，引入近似同态概念，证明线性变换器在特定条件下的收敛性，为AI模型潜在表示的结构收敛假说提供理论支持。

链接：https://arxiv.org/abs/2608.20428

机构：Dovetail Research Group（多尾研究集团）； ICC CONICET（阿根廷国家科学技术研究委员会国际合作中心）； Universidad de Buenos Aires（布宜诺斯艾利斯大学）

作者：Santiago Cifuentes

英文摘要：We study the stability of minimal representations of controlled stochastic processes (in particular, transducers) under perturbations. This question is motivated by recent experiments finding predictive-state structure in the latent representations of neural networks. We consider standard, linear and predictive transducers. We introduce notions of approximate homomorphism capturing local structural similarity between them, together with metrics comparing their induced dynamics (which we refer to as interfaces), and prove properties such as composability of the approximate homomorphisms. For standard transducers, we show that there exist simple interfaces for which there is no approximate homomorphism between the different implementations of the dynamics. In contrast, for every finite-rank interface $\mathcal I$, we prove that all minimal linear transducers implementing interfaces sufficiently close to $\mathcal I$ have an approximate homomorphism to the minimal implementation of $\mathcal I$, with error linear in the perturbation size. We prove an analogous stability result for predictive transducers under a residual metric using some mild hypothesis regarding the indistinguishability of the belief states. These results identify conditions under which canonical transducer representations are robust to perturbations, while showing that such convergence fails without additional structural restrictions. Under the assumption that these type of abstractions are embedded into the hidden layers of modern AI models, this gives some theoretical support to the hypothesis that their latent representations exhibit structural convergence.

AI 总结：本研究结合拉曼光谱与PI-AI框架，用t-SNE、K-Means等方法分析纯油及含基质油的光谱，实现食用油高精度分类，且得到的紧凑特征可大幅减少数据量，为食品质量监测提供基础。

链接：https://arxiv.org/abs/2608.20440

作者：Amrita Shaw, Chandrasekar S. N., Sai Muthukumar V., Jhinuk Gupta, Deepak L. N. Kallepalli

英文摘要：Authentication of edible oils in processed foods is important for food quality, fraud prevention, and regulatory compliance. This study establishes an integrated Raman spectroscopy and machine-learning framework that links intrinsic spectral organization, interpretable classification, and Physics-Informed Artificial Intelligence (PI-AI). Five edible oils were investigated in pure form and within a fried-potato-chip matrix using t-SNE, K-means clustering, Decision Trees, and Non-Negative Least Squares (NNLS)-based spectral decomposition. Unsupervised analyses revealed substantially stronger class organization and separability in pure oils, whereas food-matrix effects introduced pronounced spectral overlap. Decision Trees achieved 100% classification accuracy for pure oils using only four Raman variables from the original 1866-feature spectral space. These four variables, consistently identified by both pre-pruned and post-pruned models, represented only approximately 0.21% of the available spectral information while retaining perfect test-set performance. For matrix-containing samples, NNLS-based PI-AI spectral decomposition substantially improved classification by separating oil-related signatures from paper and potato contributions. Optimized post-pruned models achieved accuracies of 86.4% and 85.4% for paper-subtracted and paper-plus-potato-subtracted datasets, respectively, while reducing the number of important Raman variables to only five and four. The compact four-feature representation further reduced the data footprint by 99.44% without loss of classification accuracy. Collectively, these findings demonstrate that accurate Raman-based oil identification can be achieved through physically meaningful, highly compact, and interpretable spectral representations, providing a promising foundation for Frugal AI, Edge AI, portable sensing, and embedded food-quality monitoring.

AI 总结：本文提出一种摊销框架，通过优化对数得分学习核密度估计的带宽映射，其选择器泛化性强且性能优于经典方法，可直接应用于未知密度的有限样本。

链接：https://arxiv.org/abs/2608.20445

机构：School of Mathematics, East China University of Science and Technology（华东理工大学数学学院）； Durham University（杜伦大学）； The London School of Economics and Political Science（伦敦政治经济学院）

作者：Junyi Liang, Hailiang Du

英文摘要：Kernel density estimation converts finite samples into probability densities, but its performance depends critically on bandwidth selection. Classical selectors prescribe the sample-to-bandwidth rule analytically or asymptotically, or solve a new optimization for each sample. An amortized framework is proposed that instead learns this mapping across a distribution of density-estimation tasks by optimizing the logarithmic score. A truncated-and-renormalized bounded-support formulation enables stable learning across heterogeneous tasks, while affine standardization allows a selector trained on a single reference interval to transfer across bounded intervals. Experiments under Gaussian sampling, a multi-family benchmark, and randomized Gaussian-mixture training show that the amortized selector consistently and substantially outperforms Silverman's rule, the Sheather--Jones selector, and least-squares cross-validation, with especially large gains in small and heterogeneous samples. Finite Gaussian mixtures provide a generic training mechanism supported by their $L^1$ approximation property. Selectors trained in this way generalize strongly across different density structures, allowing the same trained selector to be applied directly to finite samples from unknown densities without specifying or fitting a distributional family. This combination of broad applicability and strong empirical performance makes the framework attractive for a wide range of applications in which finite samples or ensembles must be converted into continuous probability densities.

AI 总结：该研究针对单调对抗性损坏下的学习问题，证明多类分类等场景中部分问题不可学习，给出自适应添加数量为o(n)时仍可学习的紧界，还明确了不同对抗者下的错误率可达情况。

链接：https://arxiv.org/abs/2608.20480

作者：Julian Asilis, Shaddin Dughmi, Chirag Pabbaraju

英文摘要：Optimal learners are tailored to exploit the i.i.d.\ data assumption underlying the classic PAC model. What if an i.i.d.\ training sample were corrupted with correctly labeled examples drawn from an otherwise unrelated, even adversarial source? This model of learning with monotone adversarial corruptions was recently introduced by Larsen et al. (2026), who demonstrated that all known optimal binary learners suffer increased error rates in this setting, from $O(d / n)$ in the PAC model to $\Omega (d \log(n / d) / n)$ under monotone corruption. Mehrotra (2026) proved this logarithmic factor to be necessary for binary classification, but left open the consequences of corruption for more general learning settings, such as multiclass classification and partial binary concept classes. As our primary result, we demonstrate that monotone adversaries are frighteningly more powerful in each of these settings. We exhibit a learnable multiclass problem, of DS dimension only 2, that becomes altogether unlearnable under a monotone adversary, and show an analogous result for partial binary concept classes. These results are achieved by an adaptive adversary permitted to view the original i.i.d.\ training set $S$ and to insert $b < \infty$ corrupted datapoints into $S$. In the multiclass example, the adversary need only insert a linear number $b = |S| = n$ of datapoints. We complement these impossibility results by proving that every class remains learnable when the number of adaptive additions is $o(n)$, which our previous multiclass lower bound proves to be tight. We further observe that the classic multiclass error rate of $O(d_{\mathrm{DS}} / n)$ remains achievable against adaptive adversaries restricted to a known constant budget $b = O(1)$, against semi-adaptive adversaries viewing only a $p$-fraction of $S$ for $p \in (0, 1)$, and against oblivious adversaries that cannot view $S$.

AI 总结：本文针对Graph-JEPA出现的类别条件崩溃问题，通过实验诊断出退化解的问题并进行修复，发布了带可约性审计和目标门的工具包，揭示了联合嵌入预测架构的潜在缺陷。

链接：https://arxiv.org/abs/2608.20516

机构：L3S Research Centre（L3S研究中心）； Leibniz University Hannover（汉诺威莱布尼茨大学）； TIB Leibniz Information Centre for Science and Technology（TIB莱布尼茨科技信息中心）

作者：Gollam Rabby, Sören Auer

英文摘要：Joint-embedding predictive architectures are selected almost universally by linear probing and effective rank. We report a case where both read healthily while the representation carries zero usable instance information. We repair it, and a second failure appears: the repaired metric saturates on a target carrying no structural information. Our corpus is a scientific-reasoning graph over 57,903 articles, each a subgraph. A Graph-JEPA predicts one masked aspect from a subgraph's remaining aspects, attaining linear-probe accuracy 0.871 and effective rank 18-47, yet retrieval recovers 0.00 of 14.4 bits (MRR 1.9e-4 vs chance 1.99e-4, p=0.98). Three upper bounds on the same pool and code recover nearly everything (+14.28, +14.34, +14.22 bits), ruling out corpus, masking, pool, and metric as causes. We trace this to variance allocation - frozen inputs place 86.05% of variance on subgraph identity and 0.40% on aspect identity, while trained latents place 0.39% and 99.61%. This is a property of the objective's optimum: the degenerate solution is a global minimum of the coupled predictor/EMA-target objective, present already at init. A repaired configuration reaches 14.377 of 14.379 bits, above the 13.865-bit oracle; reverting the loss to regression drops it to 0.307 bits, confirming it. Yet the repair licenses nothing about reasoning: the target is reducible, since intra-subgraph edges are a deterministic function of node census. The oracle reaches 96.4% of the ceiling, and our largest effect is the learning-rate schedule, not architecture. Bits and a reasoning probe show no relation across ten cells. A data-derived target fails a quality gate - 25.96% of nodes are duplicate placeholders, and the rest is more generic than supporting evidence. Rank, probes, and metrics can all saturate on an unsupportive evaluation. We release a harness with a reducibility audit and target gate.

AI 总结：针对NVIDIA数据中心GPU缺乏公开SASS汇编器的问题，提出F2Asm系统，基于F₂线性代数学习SASS编码器，支持多款GPU，经测试可精确重汇编SASS指令。

链接：https://arxiv.org/abs/2608.20532

作者：Jiading Gai

英文摘要：NVIDIA provides a SASS disassembler but no public SASS assembler for recent data-center GPUs, limiting controlled machine-code rewriting. We present F2Asm, which learns exact 128-bit SASS encoders from paired disassembly and original CUBIN instruction words. To our knowledge, F2Asm is the first system to learn SASS instruction encoders as vector-valued affine maps over F2 and the first open-source NVIDIA SASS assembler to support Rubin SM107. F2Asm uses Gaussian elimination over F2 to incrementally build a compact basis, detect inconsistencies, and reject inputs outside the learned span. F2Asm separates target-specific control bits, relocation rules, and CUBIN metadata from its learning algorithm. We train encoders for Hopper SM90/SM90a, Blackwell SM100, and Rubin SM107 using 3,225 CUBINs from pinned NVIDIA and third-party production libraries, CUDA 13.3 packages, and CUDA 13.4 Developer Preview archives. In round-trip tests, F2Asm reassembles the disassembled SASS for each CUBIN, and all compared executable text sections match the originals exactly.

AI 总结：本文针对AI智能体工作流碳排放高的问题，提出AgentDecarbonizer碳优化器，可结合截止日期、电网碳强度等选择调度方案，在WildClawBench上最多降低57.9%碳排放。

链接：https://arxiv.org/abs/2608.20566

机构：University of Waterloo（滑铁卢大学）； University of Chicago（芝加哥大学）

作者：Leyi Yan, Shuangning Li, Sihang Liu

英文摘要：AI agents extend large language models from single prompt-response interactions to long-running, goaldirected workflows that issue many model calls, invoke tools, and interact with external environments. These workflows enable tasks such as software repair, data analysis, and experiment management, but their repeated model invocations can incur substantial carbon emissions. This paper characterizes the carbon emissions of OpenClaw agent workloads using WildClawBench, and shows that emissions depend on token consumption, context cache reuse, and the carbon intensity of the grid. Our characterization identifies deadline flexibility as an opportunity for carbon-aware execution: agent tasks can wait for lower-carbon-intensity periods or shift to lower-carbon grids. However, doing so requires handling uncertain execution time for temporal shifting and cached context recomputation during spatial shifting. We present AgentDecarbonizer, a carbon optimizer for AI agents that runs alongside OpenClaw. Given a task prompt and user-specified deadline, AgentDecarbonizer conservatively estimates task duration and selects deadline-feasible execution schedules, while accounting for cache recomputation overhead during spatial shifting. Evaluated on WildClawBench workloads with 60 agent tasks across four grids, AgentDecarbonizer reduces carbon emissions by up to 57.9 % compared with a carbon-agnostic baseline and by up to 37.5 % compared with a baseline that selects the carbon-optimal grid at task start time.

AI 总结：该研究针对一维二次函数设定，分析未修正Adam的稳定边缘动力学，证明其向特定稳定阈值的恢复倾向，同时指出其寻边机制失效的情况，为Adam的稳定边缘现象提供了动力学解释并揭示其局限性。

链接：https://arxiv.org/abs/2608.20638

机构：School of Engineering and Applied Sciences, Harvard University（哈佛大学工程与应用科学学院）

作者：Yiman Fong, Heng Yang

英文摘要：The edge-of-stability (EoS) phenomenon of Adam has been widely observed, while its underlying dynamical mechanism is not yet fully understood. We study uncorrected Adam on a one-dimensional quadratic, a clean setting where constant curvature isolates the optimizer-induced dynamics behind the EoS. We characterize the resulting dynamics across the parameter space. In broad regimes, we prove that Adam exhibits a restoring tendency toward its frozen stability threshold $2(1+\beta_1)/[\eta(1-\beta_1)]$. We also identify settings in which this edge-seeking mechanism breaks down, including strictly subcritical periodic orbits and specially tuned trajectories that converge to the optimum while remaining uniformly supercritical. These results give a concrete dynamical explanation for Adam's EoS in a setting free of evolving loss geometry, while also exposing its limitations.

AI 总结：本文针对开放世界未标记污染下半监督学习的鲁棒性评估问题，提出C-Score框架，通过多维度指标揭示干净准确率无法检测的隐藏退化，验证了内部诊断信号的必要性。

链接：https://arxiv.org/abs/2608.20667

作者：Tsao-Lun Chen, Chi-Cheng Fu, Han-Yi E. Chou, Shun-Feng Su

英文摘要： Pseudo-label-based semi-supervised learning has achieved strong performance due to its simplicity and scalability. However, it is typically developed under a closed-world assumption that unlabeled data are drawn from the same distribution as labeled data. In practical deployment, unlabeled data are often collected from open environments and may contain OOD samples. Under such contamination, OOD samples may still receive high-confidence predictions and be incorporated into training as if they were valid target examples. This creates an important evaluation problem: clean in-distribution test accuracy may appear stable even when the internal learning dynamics of SSL have already deteriorated. To address this issue, we study hidden collapse in pseudo-label-based SSL under open-world unlabeled contamination from a diagnostic evaluation perspective. We present C-Score, a compact framework that evaluates training behavior in three complementary spaces: prediction, feature representation, and optimization. C-Score includes PLE and CCI for unlabeled prediction behavior, Sem-Drift for deviation from labeled semantic anchors, and Grad-Align for the compatibility between labeled and unlabeled optimization. Experiments on CIFAR-10 and CIFAR-100 with multiple OOD sources, varying contamination ratios, and four pseudo-label-based SSL algorithms show that C-Score metrics reveal hidden degradation that clean accuracy alone fails to detect: under SVHN contamination, CCI rises over 280% while best-accuracy remains within 3% of the uncontaminated baseline; near-OOD sources (CIFAR-100, STL-10) cause up to 14.9% accuracy collapse (FlexMatch, r=0.5). The results suggest that clean accuracy alone is insufficient for evaluating SSL robustness in open-world environments, and that internal diagnostic signals are necessary for more reliable robustness assessment under unlabeled contamination.

AI 总结：该研究针对ReduNet参数存储量大的问题，提出轻量型自适应LA-ReduNet，采用超球面流形学习和自适应步长，仅需约1/29的参数存储即可使MCR²目标稳定，且保持相当分类准确率。

链接：https://arxiv.org/abs/2608.20668

机构：Southwest Jiaotong University（西南交通大学）

作者：Zhenglin Huang, Qifa Yan, Bin Dai, Xiaohu Tang

英文摘要：In recent years, a white-box neural network called ReduNet has been proposed, which employs the maximal coding rate reduction (MCR$^2$) principle to transform raw data into low-dimensional discriminative features via a forward layer-wise construction process. Unlike traditional deep networks that rely on backpropagation, ReduNet explicitly derives the parameters of each layer from the features of its preceding layer, offering a mathematically interpretable paradigm. However, this layer-wise construction often requires a large number of layers for the MCR$^2$ objective to reach a stable value, which increases the parameter storage of the unfolded module. To address this issue, we propose LA-ReduNet, a lightweight adaptive architecture that refines the layer-wise update rule and enables discriminative feature representations to be obtained with substantially fewer unfolded layers. Specifically, LA-ReduNet employs hyperspherical manifold learning and adaptive step sizes, thereby reducing by an order of magnitude the number of layers required for the MCR$^2$ objective to reach a stable value. Simulation results demonstrate that, while maintaining comparable classification accuracy, LA-ReduNet requires significantly fewer layers for the MCR$^2$ objective to reach a stable value. Remarkably, under the considered experimental settings, LA-ReduNet requires only approximately $1/29$ of the parameter storage of the unfolded ReduNet module for the MCR$^2$ objective to reach a stable value.

AI 总结：本研究针对长尾半监督学习的标签分布失衡与伪标签噪声问题，提出高斯桥一致性框架及BridgeMix策略，经基准实验验证可有效提升长尾类性能且具备良好可扩展性。

链接：https://arxiv.org/abs/2608.20710

机构：University of Warwick（华威大学）； Wuhan University（武汉大学）； Emory University（埃默里大学）； Peking University（北京大学）； Zhejiang University（浙江大学）

作者：Hongyang He, Xinyuan Song, Yan Zhong, Daizong Liu, Yanbin Li, Yang-fan He, Wenqiao Zhang

英文摘要：Real-world semi-supervised learning (SSL) often encounters significant challenges with long-tailed label distributions and noisy pseudo-labels, which hinder generalization and amplify confirmation bias. In this work, we introduce a novel framework, Gaussian Bridge Consistency (GBC), to address these challenges by constructing semantic interpolation paths between unlabeled samples and high-quality class anchors. Our method maintains a dynamic Prototype Atlas that stores a diverse and evolving set of labeled and pseudo-labeled exemplars per class. For each unlabeled instance, GBC forms a class-conditional Gaussian Feature Bridge in the latent space, enabling the student model to traverse a smooth trajectory from uncertain predictions to reliable class prototypes. A bridge consistency loss is applied along this path to enforce alignment with a geometrically interpolated target distribution. Furthermore, we propose BridgeMix, a confidence-aware feature mixing strategy that interpolates both sample and anchor pairs to amplify cross-sample generalization. Extensive experiments on CIFAR10-LT and ImageNet-LT (USB benchmarks) validate the robustness and effectiveness of GBC under realistic long-tailed SSL settings, consistently improving long tail-class performance without sacrificing scalability.

AI 总结：该研究针对含有限坐标观测的无限维输入浅层神经模型，构建了构造性近似与学习保证，通过完全修正贪心方法实现经验回归的统计复杂度与输入分辨率一致，并经合成实验验证了相关参数范围。

链接：https://arxiv.org/abs/2608.20812

机构：Universidad Tecnológica Atlántico Mediterráneo – UTAMED（大西洋地中海科技大学（UTAMED））； Universidad Cardenal Herrera-CEU, CEU Universities（红衣主教埃雷拉-CEU大学（CEU大学））； CUNEF Universidad（库内夫大学）

作者：Pablo M. Berná, Antonio Falcó, Diego Mondéjar

英文摘要： We develop constructive approximation and learning guarantees for shallow neural models with infinite-dimensional inputs observed through finitely many coordinates. The analysis is based on a parameter-normalized neural dictionary and its associated weighted variation class. Within this class, the approximation error separates into a distribution-dependent coordinate-truncation term and a greedy finite-width term. For empirical regression, a fully-corrective greedy procedure yields population guarantees whose statistical complexity is uniform in the retained input resolution. The same framework extends to Hilbert-valued responses without an explicit dependence on the output dimension. The dimension-free statements are statistical, not computational: selecting a new neuron still requires solving a nonconvex parameter-search problem. The quasi-Polish construction underlying recent infinite-dimensional universal approximation results provides a motivating example, and synthetic experiments illustrate the predicted resolution, width, and sample-size regimes.

AI 总结：该研究针对大型扩散Transformer的Muon优化器，提出周期性行级Muon，在保留其生成质量优势的同时，大幅降低训练的计算、通信开销与时间。

链接：https://arxiv.org/abs/2608.20818

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

作者：Chenghao Li, Xiao Han, Xinxin Huang, Wei Liu, Boyang Li, Bing Xiao, Heran Zhang, Juanma Perez Rua, Ke Xu, Kangning Liu, Linjun Kuang, Na Li, Tan Wang, Tian Xie, Wei Peng, Yang Pei, Yifan Xu, Yuanhao Zhai, Yuwei Lin, Zhe Wang, Zihao He, Daniel Li, Junbiao Tang, Ziyang Jiang, Dake Chen

英文摘要：The matrix-aware optimizer Muon improves large model training by balancing updates across singular directions, yet its scaling behavior and end-to-end efficiency on large Diffusion Transformers (DiTs) remain unclear. We first establish Muon's scaling behavior on DiTs from 1.3B to 15B parameters, showing that its optimization and generative quality advantages over AdamW persist across model scales. However, at scale, the 5-step Newton--Schulz iteration (NS5) performed at every optimization step, together with full-momentum materialization, introduces substantial computation and communication overhead that can offset Muon's step-efficiency advantage. We introduce \emph{Periodic Row-wise Muon}, which performs a full NS5 spectral update once every \(K\) steps and applies a low compute and communication cost row-wise constrained update based on the current momentum at the remaining steps. We further co-design a distributed implementation that operates directly on sharded momentum during non-refresh steps and accelerates spectral refreshes through bucketed all-gather and communication--computation overlap. Across all scales, Muon improves the best observed generative quality over AdamW by 12.9--19.1\%. Compared with vanilla Muon, Periodic Row-wise Muon remains within 0.5\% in best generative quality on the 1.3B--4B models and improves it by 4.5\% at 9B. It reduces optimizer time by 46.9--54.3\%, end-to-end step time by 15.7--24.3\%, and logical communication volume by 66.7\%, while reaching its respective best generative quality with 33.7--64.8\% less active training time. These results show that Periodic Row-wise Muon preserves Muon's generative quality advantage while translating it into end-to-end training efficiency for large DiTs.

AI 总结：CellFill方法通过在量化决策单元内写入残差实现量化大语言模型的比特一致、可撤销更新，实验显示其在召回率接近基准的同时降低跨域遗忘，且可迁移至27B混合线性注意力模型。

链接：https://arxiv.org/abs/2608.20873

机构：Institute for Frontier Interdisciplinary Research in Health Sciences and Technology, Sun Yat-sen University（中山大学健康科学与技术前沿交叉研究院）； Guangdong Engineering Research Center of Medical Artificial Intelligence Multimodal System（广东省医学人工智能多模态系统工程研究中心）； Sun Yat-sen University（中山大学）； School of Business, Sun Yat-sen University（中山大学商学院）； School of Computer Science, China University of Geosciences (Wuhan)（中国地质大学（武汉）计算机学院）； School of Public Health, Sun Yat-sen University（中山大学公共卫生学院）； Hospital of Stomatology, Sun Yat-sen University（中山大学口腔医院）； Big Data and Artificial Intelligence Center, The Third Affiliated Hospital of Sun Yat-sen University（中山大学附属第三医院大数据与人工智能中心）

作者：Zifeng Liu, Zhiyong Du, Yaxin Lu, Yiming Mao, Zhenhe Wang, Wenqi Shi, Zhengkun Jing

英文摘要：Every way of teaching a deployed language model something new -- full fine-tuning, adapter merging, model editing -- replaces the released checkpoint, and with it every evaluation and cache that referred to those exact bits. We instead learn inside the dequantization gap: with the integer codes and scales of a 4-bit release frozen, new knowledge is written only into the per-weight residual that lives strictly inside each quantization decision cell. Re-quantization then returns the released artifact bit-for-bit, a machine-checkable guarantee; updates are exactly revocable by dropping the residual; and drift is bounded. We give six propositions and three training paths, including CellFill, a bounded reparameterization that makes invariance structural rather than enforced. Exact invariance turns out to be nearly free: across three paired seeds the constrained dense path matches an unconstrained reference whose weights provably escape the artifact (58.9 vs 59.3 percent fact recall; paired difference -0.5 points, 95% CI [-5.0,+4.0]), and is better on held-out cross-domain perplexity. Against the natural null hypothesis -- serving the same update as an unmerged adapter -- projecting into the cells reduces cross-domain forgetting in every run that converged, and a diverged control shows the boundary: projection is a trust region, not a repair. What no method escapes is the cost of knowledge itself, and the apparent free lunch of in-domain perplexity improving past the anchor is an artifact of rehearsal sharing a corpus with the metric. Methods differ threefold at matched rehearsal in knowledge bought per point of cross-domain perplexity, a ranking that is not the recall ranking. The method transfers to a 27B hybrid linear-attention model (2.4e10 constrained weights, verified bit-identical), where matched recall costs about half as much cross-domain perplexity as at 1.7B.

AI 总结：该研究定义原子生成事实并构建生成事实图，基于nanoGPT等建立训练-学习统一动力学，二阶预测器在四转换上获91.43%准确率等，还将推理确立为训练-学习动力学的冻结投影。

链接：https://arxiv.org/abs/2608.20965

作者：Mian Wang

英文摘要： We define an atomic generation fact f=(u,tau,omega,z;rho), recording the origin, realized transformation, concrete occurrence, generated result and relation role. Compiled into a Generation-Fact Graph (GFG), these facts provide an AI-native, compilable scientific fact substrate preserving generation histories. We establish a GFG-based recursive scientific process in which analysis, intervention, replay and validation form facts for later cycles. Using nanoGPT, we establish unified training-learning dynamics. Training is the evolution of a parameter-optimizer system with state and memory: each actual training action enters the receiving state and produces a finite-amplitude nonlinear functional response conditioned by that state and target-specific update geometry. Learning is the persistent reorganization of distributed functional support by these responses; capability formation, maintenance, decline or recovery becomes observable when target-specific states are evaluated against their readout boundaries. Three primary coordinates - target-boundary state, target-specific update geometry and parameter-Adam receiving state - yield a second-order predictor operating before post-update outputs are read. On held-out runs, it achieved 91.43% accuracy and 91.49% macro-averaged recall across four transitions. We further establish inference as a frozen projection of training-learning dynamics. Component gating and rollback show causal recruitment and non-additive combination of query-conditioned support formed during training, deriving organizational conditions realized by Attention. Controlled feedback indicates possible double-edged reinforcement effects. ResNet/CIFAR-100 and diffusion/CIFAR-10 experiments confirm receiving-state-conditioned responses, persistent support reorganization and frozen inference projection beyond nanoGPT.

AI 总结：本文针对带文本属性图的图基础模型，提出STAG隐蔽特洛伊木马攻击框架，协调图触发器生成器与文本软提示实现后门攻击，经多数据集实验验证其有效性与隐蔽性。

链接：https://arxiv.org/abs/2608.20991

机构：The Pennsylvania State University（宾夕法尼亚州立大学）； Amazon（亚马逊）

作者：Minhua Lin, Zhicheng Gao, Yilong Wang, Hanqing Lu, Xiang Zhang, Suhang Wang

英文摘要：Graph Foundation Models (GFMs) on text-attributed graphs (TAGs) align graph representations with language semantics to support transferable graph learning. Despite these advantages, the backdoor vulnerability of GFMs on TAGs remains insufficiently understood, especially under graph-language alignment, where graph and text representations are trained to constrain each other in a shared semantic space. Existing backdoor attacks mainly target either the graph side or the text side, treating the two modalities independently. This makes direct adaptation ineffective: graph-only triggers can be constrained by clean text semantics, while text-only triggers alter the language view but do not directly shift the graph representation being aligned and scored. TAGs also impose a stealth challenge because triggers are exposed as both node text and local graph structure, making incoherent trigger attributes or anomalous subgraphs easy to inspect or filter. In this paper, we propose STAG, a stealthy trojan attack framework designed for the graph-language alignment interface of GFMs on TAGs. STAG coordinates a graph-trigger generator with a text-side soft prompt so that trigger-attached graph representations and triggered text representations move toward the same target-class text region. To address TAG-specific stealthiness, STAG realizes trigger nodes as readable text through candidate retrieval and regularizes the trigger-attached subgraph so that its local structure remains close to the original subgraph. Extensive experiments on multiple TAG datasets and representative GFMs demonstrate the effectiveness and stealthiness of STAG. Our code is available at this https URL.

AI 总结：本研究提出RODE优化器，通过解耦径向与方向分量的更新规则，在语言建模、图像分类等任务中优于Muon变体，可降低损失与模型范数，提升优化效果。

链接：https://arxiv.org/abs/2608.21024

机构：Zhejiang University（浙江大学）

作者：Guoxiang Xu, Bince Qu, Qi Sun, Cheng Zhuo

英文摘要：Modern neural network training increasingly uses matrix-aware optimizers, yet their conditioned matrix step is typically added directly to the weight, jointly changing its norm and direction. This interaction matters because the current norm determines angular motion, while directional learning can drive norm growth and thereby alter later steps. We introduce RODE, which gives the radial and directional components separate update rules and step sizes. RODE explicitly updates the matrix Frobenius norm through a scalar radial rule, while its directional channel performs Newton--Schulz-conditioned updates in the tangent space. Controlled GPT-2 interventions show gains from both direct norm control and RODE's directional update. Across two language-modeling and two image-classification tasks, RODE outperforms both Muon variants in every direct comparison and ends with lower full-model norms. At 1.5B scale, using the learning rate transferred directly from the Qwen2-style LM sweep, RODE lowers loss from 4.145 to 3.346 and final global norm from 11964 to 2183 relative to Muon RMS, with fixed-radius RODE improving further. For Qwen3.5-9B full-parameter fine-tuning, all six optimizers use the same tuning budget and the same formal-training and evaluation settings; RODE outperforms both Muon variants on all four evaluation tasks and attains the highest mean on GSM8K and MATH-500. Thus, decoupling radial and directional dynamics offers a more effective and controllable approach to matrix optimization.

AI 总结：本研究提出无训练补偿框架COEC，将其应用于Llama-3等模型的结构化剪枝，可提升剪枝后模型的困惑度与零样本准确率，在高稀疏度下增益更显著。

链接：https://arxiv.org/abs/2608.21142

作者：Peiqi Yu, Nam Ling, Wei Wang, Wei Jiang

英文摘要： Structured pruning reduces the size and inference cost of large language models (LLMs) by removing weight columns, but the resulting output error can degrade accuracy. Existing training-free compensation methods use an additive bias or a single orthogonal rotation on the output side of the retained weight. These corrections leave its input singular frame unchanged and therefore limit how the retained weight can adapt after column removal. We propose COEC (Calibrated Orthogonal-Equivalence Compensation), a training-free compensation framework that applies alternating left and right orthogonal rotations to the retained weight. The right rotation is optimized on a reduced Stiefel manifold, while singular values are rescaled using generalized cross-validation to select the regularization strength for each layer. COEC further tempers the calibration Gram matrix to reduce the dominance of high-energy activation directions and introduces an alignment penalty that preserves the geometric relation between adjacent attention this http URL components use second-order statistics from a small calibration set and require neither backpropagation through the LLM nor retraining of the model parameters. COEC is independent of the column pruning criterion and can be applied to multiple structured pruning methods. Experiments on the Llama-3, Llama-3.1, and Qwen2.5 model families across multiple structured sparsity levels show that COEC improves perplexity on every model and zero-shot accuracy in most settings over existing compensation methods, with larger gains at higher sparsity. These results show that post-pruning compensation can recover part of the performance lost to column removal.

AI 总结：针对临床叙事难以构建外科手术概率规划模型的问题，提出神经符号框架NSPIN，结合预训练LLM从2660份阑尾切除术记录生成的模型可泛化且符合临床实践。

链接：https://arxiv.org/abs/2608.21186

作者：Ranveer Singh, Saurabh Mathur, Michael Skinner, Prasad Tadepalli, Kristian Kersting, Sriraam Natarajan

英文摘要：Surgical procedures such as laparoscopic appendectomy are complex, high-stakes processes, yet formalizing their workflows for decision support remains a significant challenge. Inducing probabilistic planning domain models in this setting is particularly difficult due to the lack of structured event data and the prevalence of implicit actions in clinical narratives, which neither empirical symbolic methods nor Large Language Models (LLMs) can adequately address on their own. We introduce NSPIN, a neurosymbolic framework for inducing probabilistic planning domain models from unstructured clinical narratives. Our method extracts and imputes structured event sequences from raw text using a pretrained LLM, then induces a PPDDL model and refines its preconditions with LLM-proposed revisions, guided by empirical validation. We evaluate the approach on 2,660 laparoscopic appendectomy notes written by 9 surgeons. NSPIN yields models that generalize to unseen notes, and expert clinical review indicates its induced knowledge is largely consistent with surgical practice.

AI 总结：针对生理时间序列补全中通用模型的不足，提出CAIR两阶段框架，在多数据集多缺失机制下均实现最优性能，且能同时兼顾重建误差与临床负担指标。

链接：https://arxiv.org/abs/2608.21207

作者：Yu-Chao Huang, Haochen Zhang, Nicholas Konz, Tianlong Chen

英文摘要：Imputing physiological time series (arterial blood pressure, blood glucose, etc.) is essential for addressing the missingness that pervades clinical data. Yet modern imputation methods perform poorly in this domain: a recent benchmark found that simple linear interpolation outperformed every learned imputer on real-world clinical signals with realistic gaps. We show that this reflects two properties of physiological missingness that generic imputers ignore: gaps may occur when the signal is clinically extreme rather than typical, and gap lengths can easily span orders of magnitude. To this end, we introduce Curriculum-Aware Interpolate-then-Refine (CAIR), a two-stage framework for physiological time-series imputation. Our key motivation is to learn a coarse base curve and then repeatedly correct it toward physiological realism, rather than predict a gap in a single pass. Consequently, CAIR couples a bidirectional-GRU interpolator with a Transformer refiner that corrects its own estimate over three successive passes, trained jointly under a broad, signal-agnostic random-gap curriculum. We evaluate imputers stratified by gap length and missingness mechanism (MCAR, MAR, NMAR) rather than by a single average, and CAIR is the most accurate under every mechanism on continuous glucose monitoring (AI-READI) and arterial pressure in intensive care (MIMIC-III). Its margin over the strongest baseline grows with difficulty, from 9% under MCAR to 19% under value-dependent dropout, where generic learned imputers are weakest. We further show low reconstruction error alone does not recover the burden metrics clinicians act on: interpolants matching CAIR's error fail to preserve those metrics, imputers that recover them are far less accurate, and CAIR alone ranks among the best on both axes.

AI 总结：该研究针对测试时训练难以平衡表达性与效率的问题，提出E²-TTT方法，通过推导闭式状态转移实现并行分块训练，在语言建模、上下文检索及长上下文外推任务中表现优异，兼顾性能与训练效率。

链接：https://arxiv.org/abs/2608.21308

机构：Fraunhofer IOSB（弗劳恩霍夫IOSB研究所）； National University of Singapore（新加坡国立大学）； Lamarr Institute for Machine Learning and Artificial Intelligence（拉马尔机器学习与人工智能研究所）； University of Bonn（波恩大学）； Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院）

作者：Zeyun Zhong, Joya Chen, Manuel Martin, Frederik Diederichs, Juergen Gall, Juergen Beyerer

英文摘要： Test-Time Training (TTT) enables long-context processing via continuous weight updates during inference, but current methods struggle to balance the expressivity of per-token update dynamics with the hardware efficiency of chunk-wise approximations. We propose E$^2$-TTT (Expressive and Efficient TTT) to bridge this gap. Under the standard approximation of taking gradients at the chunk-start weights, we derive a closed-form state transition that exactly reproduces the chunk-end fast-weight and momentum states of the per-token recurrence. This enables fully parallelized chunk-level training while preserving the temporal structure of the update rule that prior chunk-wise methods discard. We validate E$^2$-TTT by training models up to 1.3B parameters from scratch. It performs on par with previous TTT and hybrid attention baselines in language modeling while outperforming them on in-context retrieval. Its advantage is most pronounced in length extrapolation: on the standard ``Needle in a Haystack'' passkey test, it retains over 90% accuracy at $8\times$ the training context length. Meanwhile, E$^2$-TTT can match the training throughput of efficient chunk-wise methods, demonstrating that it effectively reconciles expressivity with efficiency. The code is available at this https URL.

AI 总结：针对AECOPD症状变化快、现有模型延迟高的问题，本文提出基于时间感知Transformer的预测模型，利用家用呼吸机数据实现更及时的AECOPD预测，性能优于传统方法。

链接：https://arxiv.org/abs/2608.21324

作者：Weihao Qu, Ling Zheng, Dongyang Wang, Jiacun Wang, Haowen Pan

英文摘要：The rapid symptom change of Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) makes it critical to have time-sensitive prediction models. However, most current machine learning models studying AECOPD use clinical and laboratory data, which will inevitably cause latency. To ensure timely detection of AECOPD and minimize latency, this paper focuses on home monitoring scenarios where only respiratory data from daily-use ventilators is available. We introduce a Time-Aware transformer-based AECOPD prediction model, which generates meaningful patient representations using the Time-Aware transformer to capture the symptoms and their temporal progression in ventilator data. Our experimental results demonstrate that our Time-Aware transformer-based approach outperforms traditional methods in multiple classification tasks, highlighting its potential to enhance AECOPD prediction accuracy.

AI 总结：本文针对短期定价面板的跨设计不确定性，通过合成数据模拟，明确其占估计误差方差97.6%，给出经验关系，发现Paule-Mandel方差成分可提升覆盖率，为数据生成过程设计提供方向。

链接：https://arxiv.org/abs/2608.21334

机构：Universidad de Huelva（韦尔瓦大学）

作者：Pedro Cadahia Delgado

英文摘要：Short observational pricing panels can contain many observations while offering only a small number of distinct price movements. This paper studies the inferential consequences of that distinction in a synthetic data-generating process calibrated to a sparse pricing regime. We separate uncertainty conditional on a realised price trajectory from variation in estimation error across alternative trajectories generated by the same pricing process. In the baseline simulations, the latter component accounts for 97.6% of the variance of estimation error for the gradient-boosted specification. Within-panel resampling procedures use the information of one realised trajectory and do not identify this across-design component. Three results organise the analysis. First, across-design dispersion is well described by the empirical relation sigma_hat approx 0.182 V^(-0.271), where V equals moves times magnitude squared. Second, adding regions sharing a common price path reduces outcome noise but does not create independent price trajectories; conversely, averaging across units with independent design-specific errors reduces dispersion at the standard square root rate. Third, a Paule-Mandel variance component estimated across independently priced units substantially increases empirical coverage in homogeneous simulations, from 0.469 to 0.931. The broader implication is a shift toward designing data-generating processes that create independent identifying variation rather than relying solely on fixed passive panels.

AI 总结：本研究针对自精化流水线开展分阶段模型规模研究，发现生成、修订阶段需更大模型，批判阶段对规模不敏感，为多阶段语言模型系统的非对称容量分配提供了实用指导。

链接：https://arxiv.org/abs/2608.21345

机构：Drexel University（德雷塞尔大学）； University of California Irvine（加州大学欧文分校）

作者：Zhuoyi Yang, Ian G. Harris, Salar Hashemitaheri, Cassie Huang, Yuangang Li, Hyunwoo Oh, Paul Dourish, Tony Givargis, Mohsen Imani, Li Zhang

英文摘要：Self-refinement, typically structured as generation, critique, and revision, is a widely adopted paradigm for improving LLM generation and serves as a core mechanism in many LLM agents. While the three stages involve different cognitive demands, most existing approaches conveniently treat the model size as an implementation detail rather than a subject of study, which may lead to a waste of resources. Little work has systematically examined how model size affects each stage or whether effective self-refinement requires equally capable models for generation, critique, and revision. We present the first stage-wise model size study of the self-refinement pipeline on 5 benchmarks from different domains using 6 model sizes of Qwen3 and 4 model sizes of Gemma 3. We conclude that larger generators and refiners generally improve the pipeline, whereas an undersized refiner can even harm performance. Second, performance is highly insensitive to the size of the critic, although including even a small critic consistently outperforms omitting critique altogether. Our findings demonstrate that model capacity should not be allocated uniformly across self-refinement pipelines. Instead, different stages exhibit distinct size scaling characteristics, providing practical guidance for designing more computationally efficient multi-stage language model systems.