机器学习学术速递[11.6]

AI 总结：提出S³-MoE服务器端稀疏专家分配框架，解决现有联邦MoE方法无法保证恢复客户端特定相关专家的问题，在CIFAR-100实验中，仅加载5个专家的S³-MoE比Full-20 FedAvg准确率高1.31个百分点。

链接：https://arxiv.org/abs/2611.07093

机构：The Ohio State University（俄亥俄州立大学）

作者：Xue Zheng, Aylin Yener

英文摘要：Large language models (LLMs) are increasingly recognized for their remarkable capabilities; federated learning (FL) offers a promising framework for collaboratively fine-tuning LLMs on edge devices while achieving data privacy. However, edge devices are inherently heterogeneous: clients may exhibit diverse data distributions and resource constraints, including limited memory and computation budgets. Recent federated Mixture-of-Experts (MoE) approaches address these challenges by using sparse expert activation or partial expert loading. However, existing partial-expert methods rely on heuristic full-pool observations or prioritize system-wide load balance, without guarantees for recovering client-specific relevant experts from partial observations. We propose $S^3$-MoE, a server-side sparse expert assignment framework that estimates client-specific expert relevance from partial observations and refines assignments through an exploit--explore mechanism. We also establish relevance-estimation error and exact Top-$K$ recovery guarantees. Experiments using a CNN-based MoE on CIFAR-100 show that $S^3$-MoE recovers task-dependent client--expert preferences and reduces conflicts caused by correlated expert selection across clients. Compared with Full-20 FedAvg, which gives each client access to all 20 experts for local Top-5 routing, $S^3$-MoE achieves 1.31 percentage points higher mean final accuracy while loading only $5$ experts per client.

AI 总结：本文对比EEG预训练后的骨干感知适配器与容量匹配的逐词多层感知器适配器，发现其相对性能依赖骨干、任务及训练协议，无一致优势，需开展容量匹配替代方案的对比评估。

链接：https://arxiv.org/abs/2611.07192

机构：Vanderbilt University（范德堡大学）； Yale School of Medicine（耶鲁医学院）

作者：Mingyang Jiang, Yamin Li, Fan Ma, Daniel Moyer, Hua Xu, Catie Chang

英文摘要：Electroencephalography (EEG) foundation models learn representations that combine information across channels and time. This structure motivates adapters that model the same relationships during downstream learning. However, it is not clear whether such a design outperforms other candidate adapters. Moreover, in order to arrive at a fair comparison between candidate adapters, we also need to consider their trainable parameters. Here, we ask whether a backbone-informed (native) adapter makes better use of a given capacity by comparing it against a token-wise multilayer perceptron control that transforms each feature vector independently. We approximately match their trainable parameter counts and keep the pretrained encoder fixed. Across two EEG foundation models and tasks covering emotion recognition, sleep staging, and EEG event classification, the native adapter shows no consistent advantage. For one backbone, differences between the designs are small relative to variation across runs. For the other, the token-wise control performs better on emotion recognition and EEG event classification. On sleep staging, the native adapter's advantage over the token-wise control is reduced when we also train the component that processes sequences of EEG segments. The relative performance of the two adapters depends on the backbone, task, and downstream training protocol. These findings motivate evaluating backbone-informed adapters against capacity-matched alternatives.

AI 总结：本文提出用在随机感知-行动数据上预训练的Transformer替代传统方法，降低A-life智能体神经策略蒸馏的计算成本，实验验证了不同蒸馏方法及符号空间对种群动态的影响。

链接：https://arxiv.org/abs/2611.07254

机构：University of Alberta（阿尔伯塔大学）； Filuta AI； University of British Columbia（不列颠哥伦比亚大学）； Washington University in St. Louis（圣路易斯华盛顿大学）

作者：Shuwei Wang, Vadim Bulitko, Matthew R. G. Brown, Michael Youngblood, Ramon Lawrence, Roscoe Smith, Shinichi Nakagawa, William Yeoh

英文摘要： Neural networks controlling behavior of A-life agents can evolve over multiple generations. While effective in making the agents survive, such networks are generally opaque to a human observer. Furthermore, the neural weights cannot be transferred between agents with different neural architectures. Distilling such neural networks into compact symbolic policies can help humans understand agent behavior in A-life. Additionally, it allows the agents to share their knowledge in a compact fashion, improving population-level learning. Distilling a neural policy into a symbolic policy is typically performed by searching the space of symbolic policies to find one that approximates the neural policy. However, even with a domain-specific language the space of policies can be large, necessitating incomplete stochastic search such as genetic algorithms. Yet the distillation remains computationally expensive. We propose to instead use a transformer pre-trained on readily available random percept-action data. Doing so speeds up distillation substantially. We derive the transformer architecture via a neural architecture search. Empirical results compare different distillation methods and the influence of the symbolic space on A-life population dynamics through self-reflection and teaching and learning.

AI 总结：本研究探究LLM数学运算中符号错误的定位与可操控性，发现错误符号多写入于网络后期MLP层，不同模型机制不同，定位良好的电路可推翻性差异大，证明定位不意味着可操控。

链接：https://arxiv.org/abs/2611.07478

机构：Missouri University of Science and Technology（密苏里科技大学）

作者：Shradha Agarwal, Tariq J., Deepak Rajbhar

英文摘要：When large language models do multi-step mathematics, a characteristic failure recurs: the model writes one wrong sign token mid-computation -- a cofactor sign, the minus of a recursive integration-by-parts step -- then executes the rest of the arithmetic correctly, and the answer comes out wrong. We ask where inside the network that sign is written. Across six open models spanning 14B to 675B parameters, a per-layer readout shows the wrong sign is committed late -- by MLP layers in the last quarter of the network (70-100% per model and task), a pattern present, where tested (Qwen), in the base model before instruction tuning. The mechanism varies: some models carry fixed "minus-stamp" layers that push a minus regardless of the right answer, others carry amplifiers of whichever sign was already chosen, and Qwen adds a downstream rescue layer that opposes the stamp. The circuit is causal: subtracting a single sign direction from the residual stream flips 13 of 14 held-out minus-written integration-by-parts errors on Llama-3.3-70B, with flips replicating on Qwen, Gemma, and Mistral-Small (24-70% per cell), and with representation-level specificity -- the same dose breaks correctly-minus answers while sparing correct-plus answers where the direction is surgical (0/72 on Llama), and matched random-direction and control-layer interventions are null. Core claims were preregistered with pass criteria frozen in git before held-out data was seen: 23 of 29 checks passed, all misses reported. But causality is not uniform: Phi-4 implements an equally well-localized circuit at the same late depth, yet the identical intervention flips almost nothing -- the sign is written redundantly across ~15 layers with margins several times larger (median 17.8 logits). Localization, we conclude, does not imply steerability: two models can share a circuit location yet differ enormously in how overturnable it is.

AI 总结：该研究针对哈密顿量预测的等变架构存在的表示不匹配与计算瓶颈，提出DP-Learner框架，构建MP-HAM-110K数据集，实验证明其在精度和推理效率上优于现有方法。

链接：https://arxiv.org/abs/2611.07695

机构：Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合性国家科学中心人工智能研究院）； Laboratory of Quantum Information, University of Science and Technology of China（中国科学技术大学量子信息实验室）

作者：Shi Yin, Zujian Dai, Xinyang Pan, Lixin He

英文摘要：Most existing equivariant architectures for Hamiltonian prediction propagate features in direct-sum spaces of irreducible spherical tensors, whereas the target Hamiltonian is naturally an operator block in a direct-product representation over local orbital and spin bases. This representation mismatch fragments contiguous Hamiltonian blocks, obscures direct manipulation of orbital couplings, and organizes features into irregular angular-momentum components that require numerous transformations and are poorly aligned with efficient hardware execution and memory management. To address both the representation mismatch and computational bottleneck, we propose DP-Learner, an equivariant learning framework that performs message passing directly on Hamiltonian-like operator blocks in the direct-product space. Theoretically, we propose Lie-generator-based directional embedding and invariantly weighted left--right products and contractions that preserve equivariance to SO(3), SU(2), and spatial inversions, without explicit Clebsch--Gordan decomposition or inverse reconstruction. Methodologically, we instantiate this principle as a Transformer architecture, in which Hamiltonian-like feature matrices or tensors evolve through equivariant operators in the direct-product space, while non-linear invariant scalar routing controls structured gating over matrix units, directional propagation, attention, and environment-embedded context interactions. We further construct MP-HAM-110K, a large-scale Hamiltonian dataset comprising more than 110,000 material structures covering 68 elements with both Hamiltonian and wavefunction data. Experiments on MP-HAM-110K and Materials-HAM-SOC demonstrate that DP-Learner dramatically improves Hamiltonian prediction accuracy and inference efficiency over the state-of-the-art method.

AI 总结：本文针对在线策略蒸馏（OPD）的学习动力学展开机制分析，明确其反馈循环阶段，发现离线蒸馏可提升OPD的下游准确率，为OPD的学习规律与性能影响提供了理论支撑。

链接：https://arxiv.org/abs/2611.08016

机构：University of Oxford（牛津大学）； EPFL（洛桑联邦理工学院）； City University of Hong Kong（香港城市大学）

作者：Guan Zhe Hong, Sumeet Ramesh Motwani, Liang Chen, Yangchen Pan, Junchi Yu, Jialin Yu, Philip Torr, Yi Ren

英文摘要： On-policy distillation (OPD) is emerging as an important paradigm for language-model post-training, combining dense teacher supervision with training on student-generated trajectories. This creates a feedback loop where each update changes the student and thereby the training distribution that generates subsequent updates. Yet, how this feedback shapes learning and generalization remains poorly understood. We conduct a mechanistic analysis of OPD's learning dynamics, distinguishing the two feedback loop stages: conversion of the distillation signal into parameter updates, and broadcast through which these updates change predictions across prefixes and reshape the rollout distribution. This framework helps us understand both limited student responsiveness despite substantial teacher-student disagreement, and degeneration when updates that improve local teacher agreement also increase the probability of unsuccessful continuations, exposing the student to unreliable supervision in subsequent rollouts. Finally, we show that offline distillation can prepare a student for OPD without significantly improving immediate held-out accuracy. While OPD alone may degrade generation, it can yield substantial accuracy gains after sufficient offline preparation. Together, our results connect teacher supervision, student geometry, and the evolving rollout distribution in determining the learning dynamics of OPD and its effects on downstream performance.

AI 总结：该研究针对图注意力网络的解释问题，提出带自由权重的乘积规则分离价值与注意力路径的相关性，经实验验证分离的相关性可辅助解读模型决策。

链接：https://arxiv.org/abs/2611.08088

机构：Fraunhofer Heinrich Hertz Institute（弗劳恩霍夫海因里希·赫兹研究所）； Technological University Dublin（都柏林理工大学）； Technische Universität Berlin（柏林工业大学）； BIFOLD – Berlin Institute for the Foundations of Learning and Data（BIFOLD——柏林学习与数据基础研究所）

作者：Ken Schreiber, René P. Klausen, Sebastian Lapuschkin, Wojciech Samek, Jonas Naujoks

英文摘要：Graph Attention Networks (GATs) are increasingly used in domains where scientists need to read a model's reasoning, from medicine to materials science, yet the routing they learn remains opaque to most explainers. Layer-wise Relevance Propagation (LRP) follows the model's computation, but in an attention layer the relevance reaching a node is the sum of two mechanically different quantities: its content relevance through the value path and its routing relevance through the attention path. Existing rules either drop one or add them into a single number. We write a single product rule with free weights on the two factors that recovers current approaches as special cases. The value path measures what content a node supplied against a zero baseline; the attention path measures how the node shaped what was passed relative to its neighborhood, and its sign records whether the node's features moved the gating the way the prediction demanded. On a ground-truth classification task, we show that the same node receives negative, positive and near-zero relevance under three settings of the rule. The separated relevance maps yield a hypothesis about the model: it gains by not reading that node. We give reading rules for both paths and argue that they should be interpreted separately.

AI 总结：该研究针对 MoE 专家职责不明的问题，通过在 MoE 层输出训练 SAE，揭示 MoE 专家存在细粒度可解释的专业化分工，为 MoE 模型的可解释性分析提供了新方法。

链接：https://arxiv.org/abs/2611.08089

作者：Anyuan Zhuo, Shaohua Zhang, Yuan Lin, Pinyan Lu

英文摘要：Mixture-of-Experts (MoE) models route each token to a few of many experts, yet what an individual expert is responsible for remains unsettled. Existing analyses count how often an expert is selected on domain-labeled corpora, an instrument that fixes the taxonomy in advance, stays coarse, and records selection rather than contribution. We instead study specialization in feature space by training a sparse autoencoder (SAE) on the output of an MoE layer. Since this output is a router-weighted sum of expert outputs and the SAE encoder is affine, the pre-activation of every latent decomposes exactly over the experts. Aggregated over a corpus, this gives each expert a dominant set of latents whose activation it principally drives. On layers 4, 30 and 46 of Qwen3-30B-A3B, contribution is far more concentrated than gating, most latents have only one or two dominators, and the overlap between dominant sets concentrates on a few expert pairs. Auto-interpreting these sets reveals a division of labor below coarse domains, with distinct experts for Java and front-end web code, or for geometry and for arithmetic and polynomials, while the coarse domains themselves take very unequal numbers of experts. Our SAEs and code are available at this https URL.

AI 总结：该研究提出SSoG可分离高斯算子替代密集自注意力，降低计算成本，在多Transformer设置下提升图像分类准确率，且在高分辨率推理时吞吐量更高、内存占用更低。

链接：https://arxiv.org/abs/2611.08236

机构：Universitat Autònoma de Barcelona（巴塞罗那自治大学）； Centre de Visió per Computador (CVC)（计算机视觉中心）； Centre de Recerca Matemàtica (CRM)（数学研究中心）

作者：Carlos Boned Riera, David Romero Sanchez, Oriol Ramos Terrades

英文摘要：Self-attention carries a quadratic cost in token count, which limits how far Vision Transformers can scale with input resolution. We introduce SSoG, a separable sum-of-Gaussians spatial operator that replaces dense self-attention with a learned mixture of axis-factorized Gaussian kernels, reducing the per-layer cost from $\mathcal{O}(N^2)$ to subquadratic in $N$. Unlike dot-product self-attention, whose Lipschitz constant is unbounded outside a compact input domain, SSoG with fixed kernels is globally Lipschitz continuous, which, via the Picard--Lindelöf theorem, guarantees that the continuous-depth variant has a unique solution on the whole integration interval, a global guarantee that dense-attention counterparts lack in general. On CIFAR-100 at $224\times224$, in a matched discrete-Transformer setting and in the continuous setting, SSoG improves Top-1 accuracy by 2.2--4.1 points across three model widths while using 8.8--16.6\% fewer parameters, with depth and training recipe held fixed. In a looped-Transformer setting, SSoG improves Top-1 accuracy by 5.4--8.1 points. We further probe inference-time scaling by evaluating on inputs up to $1024\times1024$ at matched batch size: in the continuous-depth (Neural ODE) architecture, SSoG reaches up to 154\% higher throughput and over 60\% lower peak memory than dense attention, with the gap widening as resolution grows, consistent with the difference in asymptotic cost. In the discrete and looped settings, SSoG also achieves over 30\% higher throughput than dense attention at this resolution. Together, these results establish SSoG as a practical, parameter-efficient, and theoretically grounded alternative to dense attention across discrete, weight-shared, and continuous-depth Transformer variants.

AI 总结：本研究提出BEEMO框架，解耦行为与神经表示学习，利用未配对行为数据提升神经解码性能，仅行为预训练即可达到大神经数据集预训练模型的效果，为低成本扩展神经解码提供了可行路径。

链接：https://arxiv.org/abs/2611.08263

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

作者： Divyansha Lachi, Vinam Arora, Shivashriganesh P. Mahato, Alexandre Andre, Eva L. Dyer

英文摘要：Recent neural decoding approaches achieve strong performance by training on large collections of paired neural-behavioral recordings, but such datasets are expensive, invasive, and difficult to scale. In contrast, behavioral data is abundant and easy to collect, raising the question of whether neural decoding can benefit from decoupling behavioral and neural representation learning. We introduce BEEMO, a flexible multi-session framework that enables training with arbitrary mixtures of paired neural-behavioral recordings and unpaired behavioral data. A behavior decoder learns movement dynamics directly from behavioral trajectories, while neural activity is incorporated through cross-attention layers when available, allowing behavioral and neural supervision to scale independently. Across multiple intracortical motor datasets and tasks in nonhuman primates, we show that incorporating unpaired behavioral data improves decoding performance, particularly when transferring to new tasks, and that adding even a small amount of behavioral data from the target task further improves transfer. By decoupling neural and behavioral data, we unlock a new dimension for scaling: unpaired behavioral data, which is substantially cheaper and easier to collect than paired neural recordings, can be used directly to improve decoding performance. We further show that behavior-only pretraining, without any neural pretraining, achieves performance comparable to models pretrained on substantially larger neural datasets. Together, our results suggest that scaling behavioral data offers a practical and cost-effective path toward neural decoding models that generalize across subjects and tasks. Code is available at this https URL.

AI 总结：本文提出典型相关神经网络（CANEs），将视图内与跨视图协方差作为关系归纳偏置，在七个数据集上优于或相当于基线方法，参数更少且协方差样本减少时性能稳定。

链接：https://arxiv.org/abs/2611.07212

机构：Delft University of Technology（代尔夫特理工大学）； Indian Institute of Science（印度科学学院）； Stanford University（斯坦福大学）

作者：Andrea Cavallo, Sundeep Prabhakar Chepuri, Charilaos Kanatsoulis, Elvin Isufi

英文摘要：Relational inductive biases such as grids or graphs allow neural networks with fewer trainable parameters and improved generalization, but require a known structure beforehand. When no structure is available, the covariance matrix provides a principled substitute, and coVariance Neural Networks (VNNs) have shown that using it yields models that are theoretically grounded in Principal Component Analysis, stable to finite-sample estimation errors and parameter-efficient. This framework, however, is limited to single-view data, whereas many applications observe the same phenomenon through multiple views, each with its own in-view covariance and a cross-view covariance linking them. To incorporate these matrices as inductive biases in a principled way, we introduce CAnonical correlation NEural networks (CANEs), architectures that use in-view and cross-view covariances as distinct relational biases to solve downstream tasks. Via spectral analysis, CANEs draw analogies with Canonical Correlation Analysis (CCA), a classical technique to project two-view data onto a correlated subspace. However, while CCA's objective is to compute shared view representations, CANEs are trained to solve a downstream task directly. We bound their output error under finite-sample covariance perturbations and show that it depends on controllable Lipschitz constants of the learned filters in place of the singular-value gap that governs the stability of CCA, thus guaranteeing better robustness. Experiments on seven datasets from brain imaging and motion sensing show that CANEs match or outperform single-view, two-view and CCA-based baselines, often with fewer parameters, and retain their performance when covariances are re-estimated from fewer samples.

AI 总结：针对低秩CMDP，提出结合LSVI-UCB与原对偶技术的无模型表示学习算法MFRLC，实现与状态空间大小无关的亚线性奖励遗憾和约束违反保证。

链接：https://arxiv.org/abs/2611.07295

机构：Emory University（埃默里大学）； Shanghai Jiao Tong University（上海交通大学）； University of Wisconsin-Madison（威斯康星大学麦迪逊分校）； Stony Brook University（石溪大学）

作者：Kaixuan Liu, Guojun Xiong, Shengpu Tang, Wanyun Si, Jian Li

英文摘要：We study representation learning in low-rank Constrained Markov Decision Processes (CMDPs), where both value and reward functions are approximated by a set of unknown representation vectors, and the transition dynamics admit a low-rank factorization. The objective is to maximize expected cumulative rewards while complying with constraints on the expected cumulative utility. To achieve this, we propose \texttt{MFRLC} (Model-Free Representation Learning for Low-Rank CMDP), a model-free algorithm that uses a primal-dual scheme to effectively balance reward regret and constraint violations. To the best of our knowledge, \texttt{MFRLC} is the first representation-learning method for low-rank CMDPs that combines the Least-Squares Value Iteration with Upper Confidence Bound (LSVI-UCB) and primal-dual techniques. \texttt{MFRLC} further enhances value function estimation with bonuses for function approximation and learns the mapping from representations to value functions directly, avoiding explicit estimation of the transition kernel. We prove that, with high probability, \texttt{MFRLC} achieves $\widetilde O(K^{3/4})$ reward regret and cumulative constraint violation over $K$ deployment rounds, independent of the size of the state space. To our knowledge, these are the first sublinear regret and violation guarantees for CMDPs with unknown features.

AI 总结：该研究提出首个惯性感知基础模型IMUDINO，通过适配DINO自监督学习范式设计带频率感知位置编码的Transformer编码器，实现跨多任务的通用惯性感知，为相关领域提供骨干模型。

链接：https://arxiv.org/abs/2611.07437

机构：Tsinghua University（清华大学）

作者：Zhoutong Ye, Chengwen Zhang, Zixuan Wang, Xufeng Jian, Zhe He, Qingyang Wan, Xutong Wang, Jiaqi Liu, Mingze Sun, Yu Mei, Chun Yu, Yuanchun Shi

英文摘要： We present IMUDINO, the first foundation model (FM) for inertial sensing to achieve competitive performance across diverse sensing tasks including human activity recognition (HAR), health, motion capture, and odometry. Prior inertial FMs often focus on one subfield of inertial sensing and sometimes underperform state-of-the-art supervised methods (e.g., in HAR), making their scope considerably narrower than FMs for vision and natural language. Moreover, they often require fixed input length and sampling rate, and generalize poorly to unseen sensor locations. To build a general-purpose foundation model for inertial sensing, we adapt the DINO self-supervised learning (SSL) paradigm to inertial data through a set of semantics-preserving motion transformations that simulate variations in motion pattern and sensor setup. For model architecture, we use a plain transformer encoder with frequency-aware positional encoding that natively supports varying input length and sampling rate. This simple setup proved remarkably effective for learning global and dense features generalizable across users, sensor setups, tasks, and, in the case of robot odometry, embodiments. We hope IMUDINO can serve as a backbone for inertial sensing and kickstart the development of general-purpose inertial FMs.

AI 总结：该研究提出EARL框架，区分表征学习中状态空间与参数空间压缩两种模式，发现二者在循环储备网络的训练中分别在不同阶段出现，揭示了不变性与任务结构的学习阶段差异。

链接：https://arxiv.org/abs/2611.07952

机构：I-X Centre for AI in Science（I-X人工智能科学中心）； Department of Electrical and Electronics Engineering（电气与电子工程系）； Imperial College London（伦敦帝国学院）

作者：Hardik Rajpal, Dan Goodman

英文摘要：Learned representations show striking invariance across biological and artificial neural networks, with different substrates and variations driven by drift, heterogeneity and initialisation. To understand this invariance, we propose Emergent Adaptive Representation Learning (EARL), a framework that distinguishes two modes of emergence in learned representations: state-space compression, where a learned representation gains predictive power over its own future beyond what the network's units can predict individually, and parameter-space compression, where networks with different micro-level parameters converge to the same representation. We formalise both as Fisher sensitivities and introduce ensemble-level information-theoretic measures for each. In a recurrent reservoir with a trained bottleneck, parameter-space compression rises within the first few epochs of training and partly independently of the task, whereas state-space compression rises about 30 epochs later and only when the network begins to learn the temporal structure of the task. These results suggest that invariance to the substrate and task-specific emergent structure are acquired in distinct phases of learning.

AI 总结：该研究针对聚合物性质预测，提出基于几何划分的多尺度分子表示学习方法，构建融合3D构象信息的多模态模型，提升了性质预测精度并揭示了表示的物理相关性。

链接：https://arxiv.org/abs/2611.07998

机构：NEC Laboratories Europe（NEC欧洲实验室）； Heidelberg University（海德堡大学）； INM – Leibniz Institute for New Materials（INM——莱布尼茨新材料研究所）； Saarland University（萨尔大学）； German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））； University of Stuttgart（斯图加特大学）

作者：Makoto Takamoto, Chaitanya Sampara, Viktor Zaverkin, Daniel Oñoro-Rubio, Mathias Niepert

英文摘要：It is well established that molecular substructures can contribute unequally to chemical and physical properties, with specific motifs or side chains often playing dominant roles. Fragment-based representations have therefore emerged as an effective strategy for incorporating chemically meaningful inductive biases, as exemplified by molecular fingerprints and fragment-based attention mechanisms. However, many existing fragmentation-based machine learning methods rely on predefined chemical rules, which may yield suboptimal fragmentation for previously unseen or chemically unfamiliar structures and often overlook the importance of multiscale molecular representations. To address these limitations, we formulate molecular fragmentation as a radius-constrained optimization problem inspired by the k-center formulation, where the number of clusters is not fixed a priori but emerges from the specified radius, and propose a geometry-based fragmentation method that partitions arbitrary molecular or material structures using largely geometric information, without relying on predefined chemical fragmentation rules. The method provides flexible control over fragment size, enabling the construction of multiscale fragment representations for scale-adaptive modeling. Building on this framework, we develop a geometric fragment-based attention mechanism that integrates multiscale structural information and incorporate it into a multimodal machine learning model combining p-SMILES representations with three-dimensional (3D) conformational information. Experiments on a benchmark polymer dataset show that incorporating 3D conformational information improves prediction accuracy for several properties over models based exclusively on p-SMILES representations. Explainable artificial intelligence analyses further provide insights into the physical relevance of the learned representations.

AI 总结：本研究针对急性缺血性脑卒中后3个月功能结局预测，结合第1天多模态MRI与临床数据，发现FLAIR-临床融合模型性能最优，可提供互补预后信息，实现可解释的高效预测。

链接：https://arxiv.org/abs/2611.08154

机构：University of Geneva（日内瓦大学）； University Hospital of Geneva（日内瓦大学医院）； University of Applied Sciences Western Switzerland (HES-SO)（瑞士西部应用科学大学（HES-SO））

作者：Sina Raeisadigh, Henning Müller, Abderrahmane Hedjoudje

英文摘要：This study evaluates the added prognostic value of day-1 multimodal MRI for predicting three-month functional outcome after acute ischemic stroke (AIS). Sixty-six AIS patients with ADC, DWI, FLAIR, and clinical data were analyzed. Three-dimensional pretrained ResNet embeddings were extracted from each MRI modality, reduced using fold-isolated principal component analysis, and combined with structured clinical variables and ADC-derived lesion biomarkers. Linear support vector machines were evaluated using eight-fold stratified group cross-validation across MRI modality combinations. Clinical-only modeling achieved strong performance (AUC = $0.899 \pm 0.151$), while MRI-only models showed limited discrimination (AUC $\approx$ $0.67$-$0.73$). Among multimodal configurations, FLAIR-clinical fusion achieved the highest mean cross-validated performance (AUC = $0.948 \pm 0.061$), followed by ADC-clinical fusion (AUC = $0.941 \pm 0.094$). Full ADC-DWI-FLAIR fusion did not consistently outperform single-modality clinical fusion. Feature-importance analysis identified age, NIHSS subscores, ADC lesion burden, and learned FLAIR representations as key contributors, while qualitative FLAIR activation maps showed responses over lesion-like regions. These findings suggest that compact day-1 FLAIR and ADC representations provide complementary prognostic information beyond clinical variables and support interpretable, data-efficient outcome prediction in AIS.

AI 总结：Octopus是单模型多智能体RL框架，通过零复制分叉、仅注意力KV挂载、打包DAG-GRPO训练器三种机制，在BrowseComp-Plus等数据集上准确率优于同类方法，训练效率和内存占用更优。

链接：https://arxiv.org/abs/2611.06956

机构：Stony Brook University（石溪大学）； University of Wisconsin–Madison（威斯康星大学麦迪逊分校）； University of Minnesota（明尼苏达大学）

作者：Tianming Sha

英文摘要：Multi-agent LLM systems explore in parallel by spawning sub-agents with isolated context windows, whereas single-agent systems append every search and verification step into one growing trajectory. However, existing multi-agent frameworks duplicate prefix KV caches across branches, exchange findings by re-prefilling text summaries, and flatten or drop cross-branch dependencies during reinforcement learning (RL). We introduce Octopus, a single-model multi-agent RL framework in which all threads share one backbone and coordinate over a GPU-resident directed acyclic graph (DAG) of KV states through five primitives (fork, read_thread, wait_threads, rewind, commit). Octopus combines three co-designed mechanisms: (1) Zero-Copy Radix Forking, which spawns branches as $\mathcal{O}(1)$ cursor views over shared physical KV pages; (2) Attention-Only KV Mounting, which mounts a sibling thread's terminal KV blocks directly into full-attention layers while bypassing linear-recurrent states (Gated DeltaNet / Mamba-2), cutting off-task hijacking by $21\times$ ($57.9\% \to 2.7\%$) while retaining $92.2\%$ of full-text mutual information at zero extra KV allocation; and (3) a Packed DAG-GRPO Trainer, which packs all episode branches into one block-sparse attention pass and assigns counterfactual branch credit via leave-one-branch-out marginal rewards. Trained end-to-end with GRPO on Qwen3.5-4B over a sparse BM25 corpus, Octopus reaches $49.6\%$ accuracy on BrowseComp-Plus ($+17.5$ points over single-agent GRPO at 143K context and $+12.5$ points over text-summary multi-agent RL) and $51.2\%$ on OpenResearcher, outperforming dense-retrieval agents up to $250\times$ larger while increasing training throughput $3.3\times$ and cutting peak KV memory $2.8\times$.

AI 总结：该研究针对有限记忆POMDP，提出采用非线性神经价值函数近似的强化学习方法，建立了有限状态MDP近似误差界与神经评论者误差界，结合策略镜像下降得到有限时间次线性后悔，相关结果经数值实验验证。

链接：https://arxiv.org/abs/2611.07069

机构：Ohio State University（俄亥俄州立大学）； University of Colorado at Boulder（科罗拉多大学博尔德分校）； Johns Hopkins University（约翰斯·霍普金斯大学）

作者：Jingjuan Huang, Zhiyao Zhang, Alvaro Velasquez, Nathaniel D. Bastian, Jia Liu

英文摘要：Reinforcement learning (RL) in Partially Observable Markov Decision Processes (POMDPs) is computationally challenging due to the curse of long-range dependence on history. Fortunately, recent studies have shown that a range of POMDPs that satisfy the filter stability assumption can be well approximated by a superstate Markov Decision Process (MDP) based on a truncated history in a finite-memory window, with a value gap that decays with the window length. However, existing finite-time convergence analyses on such superstate MDP surrogates are either limited to over-simplified linear function approximation or rely on an i.i.d. sampling oracle that is violated by POMDP trajectories, both of which severely limit the applicability of these theoretical results in practice. This motivates us to bridge the gap by investigating RL in POMDPs with nonlinear neural value function approximation. Our main contributions are threefold: i) We first establish a new finite-state MDP point-wise approximation error bound, which lays the foundation for our RL algorithm design and analysis in POMDPs; ii) By applying Temporal-Difference (TD) learning to POMDP-generated superstate transitions and targeting the Bellman fixed point of the superstate MDP, we prove a finite-time critic error bound with nonlinear neural value function approximation; and iii) By combining the neural critic function approximation error bound with the policy mirror descent framework, we obtain a finite-time sublinear regret against the optimal POMDP value. To the best of our knowledge, such a finite-time sublinear regret based on neural function approximation is the first in the literature for finite-memory POMDPs. Our numerical experiments also verify our theoretical results.

AI 总结：本研究提出用固定与自适应惩罚调控AMoD运营商，通过图多智能体强化学习建模，在曼哈顿数据实验中可降低过高定价比例、提升区域服务可用性，明确了调控目标等间的权衡。

链接：https://arxiv.org/abs/2611.07122

机构：Technical University of Munich（慕尼黑工业大学）； Technical University of Denmark（丹麦技术大学）

作者：Korbinian Kügler, Carolin Schmidt, Florian Dandl, Filipe Rodrigues

英文摘要：Profit-maximizing Autonomous Mobility-on-Demand (AMoD) operators may exploit consumer surplus through excessive prices and underserve less profitable regions by relocating vehicles away from them. We study how penalty-based regulation of an AMoD operator in a multimodal mobility system can limit surge pricing and improve regional service availability. We model regulation as an outer control loop that acts on a learning-based operator. Static regulation with fixed penalties is formulated as a Markov Decision Process for an operator that jointly controls pricing and rebalancing, and dynamic regulation as a Partially Observable Stochastic Game in which a regulator periodically adjusts region-specific penalties in closed loop. Operator and regulator policies are learned with graph-based multi-agent reinforcement learning. In experiments with real-world Manhattan data, fixed surge pricing penalties reduce the share of travelers facing excessive prices from 22% to 1%. For regional service availability, dynamic regulation improves the regulator's reward by up to 15.6% over fixed penalties of equal mean intensity. The gains in service availability are strongest when spare fleet capacity exists, whereas service deficits persist in peak periods with highly utilized fleets. The learned regulator policy responds by reducing penalties where limited fleet capacity restricts compliance. The results identify trade-offs between regulatory targets, operator profitability, and traveler utility.

AI 总结：该研究提出规定性策略学习（PPL），直接向智能体传授行为原则而非依赖奖励信号，使智能体无需奖励即可学习到更优且泛化能力更强的策略。

链接：https://arxiv.org/abs/2611.07130

机构：National University of Singapore（新加坡国立大学）

作者：Shan Yang

英文摘要：Reinforcement learning has achieved strong performance in many tasks by letting agents learn from nothing more than a reward signal. However, in more complex tasks, agents still struggle to learn good, generalizable policies, even with carefully designed rewards and abundant samples. We argue that this difficulty stems from the reward decoding burden: designers encode their knowledge of the desired behavior into rewards and penalties, and agents must infer this knowledge from long sequences of noisy scalars. To address this burden, we propose prescriptive policy learning (PPL), which directly teaches agents higher-level knowledge about the tasks, termed principles of behavior, such as how to maintain balance, how to accumulate energy, and how to land smoothly. Through appropriately selected key quantities and references, PPL can directly indicate at every step whether the agent follows the principles and to what extent. Our experiments show that agents can learn good, generalizable policies without any reward. Taught only simple heuristic principles, PPL agents outperform reward-based methods with far fewer samples. Their policies also transfer directly to harder settings and keep performing well, demonstrating their strong generalization ability. In summary, PPL provides a new approach in which rough knowledge of the desired behavior is enough for agents to learn good, generalizable policies, even when the system dynamics cannot be accurately modeled.

AI 总结：本文针对大视觉语言模型的组合性安全问题，提出无训练不对齐方法CMaG，通过分离跨模态交互项引导拒绝转顺从，在两款模型上降低组合安全评分65个百分点且保持98%以上输出连贯。

链接：https://arxiv.org/abs/2611.07216

机构：Emory University（埃默里大学）

作者：Hong kyu Lee, Kai Shu, Li Xiong

英文摘要：Large vision-language models (LVLMs) are vulnerable to compositional harm: an image and a prompt that are benign in isolation become harmful through the model's cross-modal reasoning. Recent alignment methods report strong scores on compositional safety benchmarks, but an LLM judge only measures observed behavior and cannot reveal how robustly safety is integrated into the model. We argue that un-alignment provides a complementary audit of alignment robustness. We first show that minimal supervised fine-tuning reaches an LLM-judge score comparable to a multi-stage reinforcement-learning pipeline, so similar benchmark scores cannot establish robustness by themselves. We then show that recognition and refusal of compositional harm are decoupled: nine standard direction-finding methods detect compositional harm with up to AUC=0.99, yet ablating the recovered directions either leaves refusal intact or destroys model coherence. We therefore propose CMaG, a training-free un-alignment method for compositional safety. CMaG measures refusal intention with a refusal-versus-compliance log-probability margin, takes its gradient with respect to mid-layer activations, and contrasts it against partially benign pairs to isolate the cross-modal interaction term. A single backward pass over 80 examples yields a direction along which steering monotonically converts refusal into compliance. On Qwen2.5-VL-7B and LLaVA-1.5-7B, CMaG reduces compositional safety scores by up to 65 points while over 98% of outputs remain coherent, outperforming gradient-based baselines; the direction also transfers to text-explicit harm and un-aligns models hardened by defensive training. Compositional safety can thus appear behaviorally strong while remaining fragile to targeted internal intervention. Warning: this paper contains unsafe model outputs, including references to self-harm.

AI 总结：该研究提出MAC-v2，通过带动作分块的无重构潜态世界模型，解决基于像素的长时程基于模型的价值展开难题，在多个视觉基准中优于现有离线RL基线。

链接：https://arxiv.org/abs/2611.07245

机构：UC Berkeley（加州大学伯克利分校）

作者：Kwanyoung Park, Sergey Levine

英文摘要：We study whether \emph{long-horizon model-based value expansion} can serve as an effective framework in pixel-based offline reinforcement learning (RL). Long-horizon model-based value expansion from pixels is substantially harder than state-based environments, as the agent must learn both the representation and its dynamics, and errors from either can accumulate recursively over imagined trajectories. We introduce MAC-v2, which enables long-horizon model-based expansion from pixels by training a reconstruction-free latent world model with action chunking. By learning representations from prediction and control objectives, MAC-v2 does not need to encode control-irrelevant visual details that can hinder the dynamics prediction. Action chunking further mitigates the prediction error by reducing the number of recursive model predictions required for value expansion. Across diverse visual benchmarks including visual OGBench, MMBench2, Robomimic, and LIBERO-10, MAC-v2 achieves strong performance against model-free and model-based offline RL baselines. Our analyses show that long-horizon model-based value expansion is critical for the performance of MAC-v2, and that both action chunking and latent learning objectives are essential for reliable long-horizon latent imagination.

AI 总结：该研究提出类贝叶斯模型ACB，将未达成目标作为先验约束决策，通过伪遗忘机制区分于带显式遗忘的模型，模拟显示其可减少老虎机决策的遗憾，且能预测生物对放弃选项的重新评估。

链接：https://arxiv.org/abs/2611.07270

机构：Graduate School of Advanced Science and Engineering, Tokyo Denki University（东京电机大学先进科学与工程研究院）

作者：Moto Shinriki, Yu Kono, Tatsuji Takahashi

英文摘要： Animals and humans often do not maximize; they act relative to goals (aspiration levels) that are frequently not yet met. How can a goal that is not (yet) realized organize choice? Deacon (2012) calls such absence-referenced organization \emph{ententional}, and García-Valdecasas and Deacon (2024) argue that ends are best understood as constraints rather than as particular states. We examine this idea in a minimal formal setting: Risk-sensitive Satisficing (RS), a bandit learning rule that values each action by its share of past trials multiplied by its surplus or shortfall relative to an aspiration level. We show that RS can be rewritten as a Bayesian-like estimator, Attention Competitive Bayes (ACB), in which the aspiration level is the prior mean of every action and the finite budget of trials sets how strongly evidence can override it. Two consequences follow. First, actions that are no longer sampled are partially overridden by the goal rather than by past observations: their estimates relax toward the aspiration level without any data being discarded. This ``pseudo-forgetting'' distinguishes ACB from models with an explicit forgetting rate. Second, when the aspiration is unattainable, exploration does not settle; in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss). ACB offers an analyzable model in which an unattained goal and a limited attentional resource jointly constrain behavior, and it yields testable predictions about how organisms revalue options they have abandoned.

AI 总结：针对持续强化学习的稳定性-可塑性权衡问题，本文提出系统级分解思路，引入非参数化近似估计临时价值函数，在图像任务和Craftax-Classic上优于基线，验证了该方法的有效性。

链接：https://arxiv.org/abs/2611.07327

机构：School of Computer Science（计算机科学学院）； McGill University（麦吉尔大学）； Mila（米拉研究所）； Google DeepMind（谷歌DeepMind）

作者：Nishanth Anand, Doina Precup

英文摘要：Continual Reinforcement Learning (CRL) agents struggle to adapt to new situations while retaining prior knowledge, leading to a stability-plasticity trade-off. One approach to solving this problem is by using complementary learning systems: one for slow, long-term learning and another for quick, transient adaptation. Anand & Precup (2023) instantiated this idea using permanent and transient value functions, but used the same features in both systems. Intuitively, long-term learning should rely more on parametric representations for broad generalization, while transient learning would be best served by non-parametric representations for situation-specific adaptation. In this paper, we explore this idea both theoretically and empirically. By introducing a novel non-parametric approximation for estimating the transient value function in complex tasks, we demonstrate that our method enables online learning and outperforms competitive baselines on image-based tasks and Craftax-Classic, underscoring the effectiveness of our system-level decomposition.

AI 总结：该研究针对基于决策的稀疏对抗样本攻击问题，提出RLGPA强化学习方法，通过调度支撑缩减执行器及其粒度，在CIFAR-10等数据集上显著提升了目标攻击成功率。

链接：https://arxiv.org/abs/2611.07512

机构：Huzhou College（湖州学院）； South China Normal University（华南师范大学）； Guangdong University of Education（广东第二师范学院）

作者：Chao Zhou, Shenghua He, Jiaxin Mo, Qiangguo Yu, Fengchuang Xing

英文摘要：Decision-based sparse attacks must reduce perturbation support using only the victim's predicted label. The best restoration strategy varies as support contracts: coarse removal works early, while tail states need finer, history-aware tests. We propose RLGPA, an online reinforcement learning method that schedules support-reduction executors and their granularity under a hard query budget. A masked Dueling Double DQN selects among seven stable policy--scale actions; each action produces support-relative restoration groups that are rebuilt after every macro decision. A feasibility-preserving executor commits only adversarial restorations and feeds support gain and query cost back to the planner. Under the matched top-1-label oracle, RLGPA reaches 61\% targeted ASR at sparsity 0.01 on CIFAR-10 (vs. 25\% SparseEvo, 0\% Pointwise), and 75.5\%/73.0\% on ResNet50/ViT at sparsity 0.005 (vs. 29.0\%/15.5\%). A fixed-probe test reduces median \(L_0\) from 165 to 23 after 40 images without any update. State-blind and state-shuffled controls degrade significantly, confirming that the learned policy uses attack state; the main gains come from state-conditioned executor scheduling and dynamic repartition. The source code is publicly available at this https URL.

AI 总结：针对基于人类反馈的强化学习（RLHF）训练中存在的约束违反问题，提出无奖励模型的零阶对数障碍策略梯度方法，结合保守约束估计等策略保证训练中所有策略安全可行，证明其收敛性并通过实验验证安全保障。

链接：https://arxiv.org/abs/2611.07561

机构：Purdue University（普渡大学）； UPenn（宾夕法尼亚大学）； University of Tennessee, Knoxville（田纳西大学诺克斯维尔分校）； RPI（伦斯勒理工学院）

作者：Weiqin Chen, Xinmeng Huang, Dongsheng Ding, Santiago Paternain

英文摘要：Safe reinforcement learning from human feedback (Safe-RLHF) seeks to improve helpfulness while satisfying harmlessness constraints. Existing primal-dual approaches can converge to feasible policies but do not prevent constraint violations during training. We study a stronger requirement, safe exploration for RLHF, under which every policy executed during training, including auxiliary exploration policies, remains feasible. We propose Zeroth-Order Log-Barrier Policy Gradient (ZO-LBPG), a reward-model-free method that optimizes a log-barrier objective using zeroth-order estimates derived directly from human preferences. To ensure safety under noisy feedback, we combine conservative constraint margin estimates with coupled rollouts and an adaptive step size that prevents updates from crossing constraint boundaries. We prove that, with high probability, all iterates and exploration policies remain strictly feasible, and that ZO-LBPG globally converges to an approximately optimal constrained policy. We establish a sample complexity of $\widetilde{\mathcal O}(\varepsilon^{-6}+d^3\varepsilon^{-5})$ trajectories and $\widetilde{\mathcal O}(\varepsilon^{-8}+d^5\varepsilon^{-7})$ human queries. Experiments on a synthetic CMDP and a language-model alignment task corroborate the training-time safety guarantees.

AI 总结：本研究针对部分可观测下sLLM智能体的指令-对象绑定问题，提出无训练框架BINDIntent，显著提升了同义词释义下的任务成功率，性能超24B提示式LLM。

链接：https://arxiv.org/abs/2611.07599

机构：Chung-Ang University（中央大学）； Soongsil University（崇实大学）

作者：Chanyoung Kim, Harim Song, Byoungchan Ko, Keonyoung Lee, Myungwon Lee, Dahuin Jung

英文摘要：Reliable deployment of LLM agents requires grounding diverse user expressions in their environments. Although RL-finetuned smaller large language model (sLLM) agents rival larger prompted LLMs on embodied tasks, their robustness to instruction paraphrases remains unclear. We introduce ALFWorld-Para to evaluate this robustness under controlled paraphrases, revealing 21-52% declines in success rates (SRs), predominantly driven by object synonyms. Despite retaining knowledge of synonym referents, agents often fail to take the corresponding objects even after observing them, exposing a gap between referential knowledge and object binding. Partial observability further complicates binding, as referents may remain unseen or have already been encountered under different labels. To address this challenge, we propose BINDIntent, a training-free framework that probabilistically binds referring expressions to observed labels using semantic proximity and estimated persistence against unseen candidates. Without additional training or LLM calls, BINDIntent improves synonym-paraphrase success of RL-finetuned sLLMs by 17.5-49.1 percentage points (pp) with only 0.4 ms of CPU overhead per step. It also raises GiGPO-7B's overall SR from 61.8% to 77.4%, surpassing a prompted 24B LLM. Code and benchmark are available at this https URL.

AI 总结：提出单调生存性学习（MoViL），通过单调有限时长集递推学习生存集，在安全RL基准上实现最优奖励-成本性能，可用于安全关键系统的动作验证。

链接：https://arxiv.org/abs/2611.07616

机构：Tsinghua University（清华大学）

作者：Ziqian Wang, Zhen Zhang

英文摘要：Offline safe reinforcement learning requires reliable long-horizon safety estimates from fixed data. Learning Hamilton-Jacobi (HJ) viability in this setting calls for semantic preservation, convergence, and numerical stability. To address these challenges, we propose Monotone Viability Learning (MoViL), which learns viable sets from offline transitions through monotone finite-horizon set recursion. The recursion preserves finite-horizon safety semantics and converges to the infinite-horizon viability kernel. Binary supervision maintains a fixed target scale across horizons, avoiding discount-induced numerical attenuation. We further establish sufficient conditions for recovering behavior-supported viable subsets. The learned viability estimates are used to certify actions at deployment. Dubins car experiments validate the numerical accuracy of the learned viable sets. Evaluations on safe RL benchmarks demonstrate state-of-the-art reward-cost performance. MoViL satisfies the average cost constraint on every task and achieves the lowest average cost among the compared methods. These results highlight MoViL's potential to extend long-horizon HJ viability analysis to safety-critical systems where reliable dynamics models are unavailable and online exploration is unsafe.

AI 总结：该研究构建gpu-train-debug基准，发现基于验证器的RL在分布内无明显改进，仅在分布偏移下有适度增益，SFT贡献了大部分改进，研究成本低且可复现。

链接：https://arxiv.org/abs/2611.07639

机构：Santa Clara（圣克拉拉）

作者：Amit Sagtani, Saloni Garg

英文摘要：Reinforcement learning from verifiable rewards (RLVR) is usually evaluated as an alternative to supervised fine-tuning (SFT), or as a stage that follows it, with the implicit assumption that the RL stage contributes gains of its own. We measure that contribution directly, in a controlled setting where both stages can be run to completion and compared exactly. We built gpu-train-debug, a deterministic benchmark of 1,877 synthetic GPU and LLM training failure diagnosis tasks across 30 archetypes, graded by a programmatic verifier with partial credit, and calibrated so that a mid-tier model scores in a 0.60 to 0.75 band. A pilot produced a surprise: 1,200 gold demonstrations nearly solved the benchmark (SFT 0.785 on the hard split), while 15 steps of GRPO from the base model did not move it (0.314 vs 0.311 base). We therefore amended the study, in writing and before the full run, to ask the sharper question: what does verifier-based RL add beyond demonstrations? At the pre-registered scale (60 GRPO steps, learning rate 1e-6), the answer is: no detectable improvement in-distribution (+0.22 points, 95% CI [-0.06, +0.51]), and a modest but measurable improvement under distribution shift (+0.33 points on a held-out OOD split, CI [+0.02, +0.68]). SFT accounted for most of the available improvement, while verifier-based RL produced only modest additional gains under distribution shift. The residual errors SFT leaves are mostly systematic, identical across samples, which likely reduces the within-group learning signal available to GRPO. A full 60-step cold-start GRPO arm confirms the pilot at the full horizon. The whole study cost $8.69 of compute and is fully reproducible from archived generations, frozen configs, and tagged code.

AI 总结：该研究针对有限弱通信MDPs，证明同步RVI Q学习的收敛速率为$O(1/n)$，需$O(\epsilon^{-1})$次同步迭代可得期望$\epsilon$最优策略，保证常数依赖于到达时间参数与次优差距。

链接：https://arxiv.org/abs/2611.07676

机构：Indian Institute of Science（印度科学学院）； Stanford University（斯坦福大学）； EPITA（EPITA（法国高等计算机与技术学院））

作者：Avik Kar, Siddharth Chandak, Rahul Singh, Shalabh Bhatnagar

英文摘要：We establish finite-time convergence guarantee for synchronous relative value iteration (RVI) $Q$-learning for finite weakly communicating Markov decision processes (MDPs). For stochastic Krasnosel'ski\uı--Mann iterations driven by a non-expansive, linearly regular map on a finite-dimensional normed space, we establish last-iterate mean-squared residual bounds under martingale-difference noise whose conditional second moment grows at most linearly with the squared residual. Suitable polynomial stepsizes of order $n^{-a}$ yield an $O(n^{-a})$ rate for $0

AI 总结：针对扩散语言模型块级似然难以计算的问题，提出PBPO方法，通过前缀块状态与单步替代项实现块级策略优化，在多基准测试中取得与现有方法相当的性能。

链接：https://arxiv.org/abs/2611.07783

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

作者：Katherine Tieu, Dongqi Fu, Si Zhang, Hanqing Zeng, Tianxin Wei, Chenglin Wei, Yinglong Xia, Hong Li, Jingrui He

英文摘要：Diffusion Large Language Models (dLLMs) commonly decode responses by proceeding autoregressively across blocks while iteratively denoising multiple tokens within each block. This structure makes block-level policy optimization a natural adaptation of autoregressive policy optimization. However, the exact likelihood of a completed block is intractable because it marginalizes over the intermediate states and reveal orders of the denoising process. To realize the block-level formulation, we construct a prefix-block state and use a single model evaluation to define a factorized one-step surrogate for the block likelihood. The resulting block surrogate ratio is a product of its token surrogate ratios, allowing variation at individual positions to propagate across the entire block before clipping. We therefore introduce Prefix-Block Policy Optimization (PBPO), which preserves block-level conditioning but applies surrogate importance weighting and clipping separately to each token. PBPO further diversifies repeated updates by resampling corrupted views of the prompt and revealed prefix, a design motivated by masked-dLLM pre-training. Under our analysis assumptions, block-product weighting amplifies the variance of gradient contributions relative to token-local weighting. Across two LLaDA-family initializations, six benchmarks, and three generation lengths, PBPO achieves competitive performance against existing dLLM RL methods.

AI 总结：AttnSR 是一种步级奖励增强方法，利用策略注意力推导奖励分配，在 ALFWorld 和 ScienceWorld 上均优于 PPO 基线，提升了长 horizon 智能体强化学习的性能。

链接：https://arxiv.org/abs/2611.07932

机构：Shanghai Institute of AI for Education, East China Normal University（华东师范大学上海教育人工智能研究院）； Shanghai Innovation Institute（上海创新研究院）； Hithink Research（恆大研究院）； School of Data Science and Engineering, East China Normal University（华东师范大学数据科学与工程学院）； The Hong Kong University of Science and Technology（香港科技大学）

作者：Mengsong Wu, Tianchen Zhou, Zhaochen Su, Zhenhua Liu, Zhe Cui, Aimin Zhou

英文摘要：Long-horizon LLM agents make multi-step decisions, yet outcome-reward-only reinforcement learning provides little direct supervision for individual steps. We introduce AttnSR, a step-level reward augmentation method that derives reward allocations from policy attention extracted during training replay, without a separately trained reward model or additional rollouts. AttnSR traces backward through the policy model attention graph to estimate the relevance of earlier steps to the final response. The terminal step, measured relative to a running baseline, determines whether these rewards reinforce or penalize the selected steps. In a controlled Qwen2.5-7B-Instruct study, AttnSR consistently outperforms the PPO baseline across all evaluated checkpoints on ALFWorld, raising final held-out success from 91.2% to 97.1%. The ablation experiment shows that the benefit depends on how intermediate rewards are assigned, rather than simple accumulation of step rewards over time. AttnSR also improves ScienceWorld at every checkpoint while cutting environment interactions by 29.9%. These results support policy attention as a practical prior for step-level reward allocation in long-horizon agentic RL. Code is available at this https URL.

AI 总结：本文提出决策模型引导训练（DMGT）技术，结合决策模型（DM）与强化学习（RL）的优势，在7项任务上的实验显示其性能优于RL和多数DM，还给出了DM环境与任务描述的指南。

链接：https://arxiv.org/abs/2611.07934

机构：Coventry University（考文垂大学）

作者：Muhammad Ibrahim Khan, Aniket Dixit, Faizan Ahmed, James Brusey

英文摘要：Reinforcement learning (RL) provides a framework for continuous improvement through interaction with an environment. However, RL agents struggle against tasks that offer exploration challenges and sparse rewards, because learning efficient policies from scratch is slow and sample-inefficient. In contrast, decision models (DMs), given a description of the environment and task, can perform well in sparse-reward tasks, but, as they do not learn from reward, they are not able to improve towards the optimal solution. In this paper, we present Decision Model-Guided Training (DMGT), a technique that uses a DM with a description of the environment and task to initialise an RL policy, combining the DM's ability to act without training with RL's ability to improve from reward. Instead of exploring from scratch, the agent reaches the goal and receives reward from its first episodes, and RL then improves its policy beyond what the DM achieves. We demonstrate through experiments on seven tasks that our approach outperforms RL in all seven and DM in five. In addition, we provide comprehensive guidelines, derived from ablation studies, for describing an environment and task to a DM.

AI 总结：该研究提出FewStepRL，通过变换、分解、适配三阶段将RL目标适配到少步扩散模型，在模块化代码库中实现多种现有方法的少步版本，实验表明奖励梯度方法在奖励可微时更有效，额外推理步骤无系统总体益处。

链接：https://arxiv.org/abs/2611.08240

机构：Tencent Hunyuan（腾讯混元）； Westlake University（西湖大学）； University of Chinese Academy of Sciences（中国科学院大学）

作者：Haitao Lin, Minsi Ren, Yang Li, Xingtong Ge, Zheyuan Hu, Peiyan Hu, Tailin Wu

英文摘要： Reinforcement learning (RL) has become a standard post-training method for diffusion models, but most RL objectives roll out the current policy at every iteration, and with a many-step sampler these rollouts typically dominate the cost of training. Few-step models cut this cost, yet they model an average velocity over an interval, whereas existing RL objectives are built for an instantaneous velocity with a many-step sampler. We adapt RL objectives to few-step models systematically in three phases, namely a transformation, a decomposition and an adaptation. The transformation to $x$-space expresses each objective we consider through a common policy prediction, at the clean end by default and at an explicitly marked level for transition or flow-map objectives, so that objectives stated over densities, velocities or rewards become comparable. The decomposition writes the gradient of each of these objectives as the sum of a self-regression toward a reference and a reward augmentation, so that a method reduces, at the gradient level, to a choice of anchors and weights. The adaptation computes each anchor from a few-step model at the end of its transport or along its tangent, with the instantaneous velocity recovered from the network itself or from an auxiliary network. This yields few-step versions of FlowGRPO, AWM, DiffusionNFT, DiffusionOPSD and ReFL in one modular codebase. In our experiments, reward-gradient-based methods are the most effective when the reward is differentiable and its gradient is affordable; otherwise DiffusionNFT-style methods generally reach high rewards in relatively few iterations despite a sometimes higher per-iteration cost. Extra inference steps show no systematic aggregate benefit in our experiments, although selected outputs visibly improve with two to four times as many sampling steps as in training.

AI 总结：该研究针对大规模并行安全强化学习与约束马尔可夫决策过程的不匹配问题，提出相位聚合成本估计器与约束收紧框架，实现了训练时间大幅减少且性能相当的可扩展安全强化学习方法。

链接：https://arxiv.org/abs/2611.08273

机构：Pusan National University（釜山国立大学）

作者：Joonyoung Lim, Younghwan Yoo

英文摘要：We investigate the problem of scaling safe reinforcement learning (safe RL) to massively parallel training regimes, where a large number of environments are exe- cuted with short rollout horizons. Although this setting improves data throughput and wall-clock efficiency, it creates a mismatch with constrained Markov decision processes (CMDPs), whose safety constraints are defined over full-episode cost returns. We address this mismatch by analyzing staggered environment resets as a phase-mixture distribution over short trajectory segments. This analysis charac- terizes the bias induced by phase coverage and motivates a phase-aggregated cost estimator that reconstructs episode-level cost estimates from short rollout batches. We further incorporate finite-sample uncertainty and distribution mismatch into a constraint-tightening framework, yielding a scalable Safe RL method compatible with on-policy optimization. Using an MJX-based implementation of Safety Gym, our experiments show that the proposed method achieves performance comparable to non-massively parallel implementations while substantially reducing wall-clock training time.

AI 总结：针对一般和博弈中现有对手塑造方法需获取对手信息的局限，提出DUOS方法，仅利用自身经验发现更新规则，在归一化平均得分和自对弈中均排名第一。

链接：https://arxiv.org/abs/2611.08335

机构：Harvard University（哈佛大学）； University College London（伦敦大学学院）

作者：Hong Jun Lee, Ziyan Wang, Jun Wang, Stefanos Leonardos, Yali Du

英文摘要：In general-sum games, self-interested learners can reach outcomes that are poor for everyone, which has motivated methods that shape the opponent's learning. Methods like Learning with Opponent-Learning Awareness (LOLA), which differentiates through the opponent's learning process, and Model-Free Opponent Shaping (M-FOS), which uses model-free optimisation methods for long-horizon opponent shaping, have enabled mutual cooperation and exploitation of learners. However, these methods require access to the opponent's parameters, policy, rewards or learning rule during deployment. In single-agent reinforcement learning, update rules discovered by meta-learning have achieved state-of-the-art results (e.g. DiscoRL). These rules are separate from the policy they train and read only the agent's own experience. We hereby extend this framework to general-sum settings by introducing Discovered Update Rule for Opponent Shaping (DUOS). To the best of our knowledge, this is the first method to discover an update rule for an agent in general-sum games. DUOS meta-trains the rule to maximise the agent's expected return over its lifetime, and updates an inner network using the agent's own experience. We show that our discovered rule exhibits behaviour consistent with opponent shaping without reading opponent policies or parameters, and ranks 1st in normalised mean score and 1st in self-play among tournament entrants.

AI 总结：该研究探究了监督微调（SFT）与强化学习（RL）中LoRA的谱与子空间动力学，明确组均值基线减法是两者差异的来源，据此定制的LoRA更新配置在数学推理任务上取得了最佳性能。

链接：https://arxiv.org/abs/2611.08414

机构：New York University（纽约大学）； Institute of Science and Technology Austria（奥地利科学技术研究所）

作者：Chenruo Liu, Zihan Wang, Yijun Dong, Qi Lei

英文摘要：Low-rank adaptation (LoRA) is commonly used in both supervised fine-tuning (SFT) and reinforcement learning (RL), often under similar optimizer settings. Yet, how the different dynamics of the two stages manifest in the low-rank increment that LoRA optimizes remains unclear. Tracking the singular spectrum and subspaces of LoRA updates on mathematical reasoning tasks, we find that RL concentrates the spectrum at every learning rate, while SFT expands it at moderate learning rates; moreover, RL rotates the singular subspaces much faster than SFT at small learning rates. Interpolating between SFT and RL one ingredient at a time, we pinpoint group-mean baseline subtraction as the source of both contrasts. A quadratic multi-index model updated via LoRA unveils the underlying mechanism: group-mean baseline subtraction tilts the balance between two competing forces, coherent fitting and gradient noise, toward spectral concentration and faster subspace rotation, while forcing the risk-favorable learning rate to be low. Guided by these dynamics, we tailor LoRA updates to each stage: in SFT, we flatten the spectrum of the full update to spread adaptation across singular directions, whereas in RL, we flatten the part of the update outside the current LoRA increment span to promote subspace rotation. Experiments on mathematical reasoning tasks across multiple models show that the respective configurations for SFT and RL yield the best performance among the compared methods in most evaluated settings.

AI 总结：该研究对比OPD与GRPO的计算效率，发现当教师为学生自身GRPO检查点时，OPD在数学等任务上效率更高，多教师OPD可整合GRPO专家并降低总计算量。

链接：https://arxiv.org/abs/2611.08453

机构：Salesforce AI Research（Salesforce人工智能研究部）； UIUC（伊利诺伊大学厄巴纳-香槟分校）

作者：Zeyu Leo Liu, Yang Li, Bo Pang, Emre Can Acikgoz, Xuan-Phi Nguyen, Anurag Koul, Srijan Bansal, Erik Nijkamp, Shafiq Joty, Semih Yavuz

英文摘要：On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher's per-token supervision. Does this dense signal reduce the training FLOPs needed to reach a target accuracy compared with reinforcement learning? We systematically compare vanilla reverse-KL OPD with Group Relative Policy Optimization (GRPO) across task difficulty, student and teacher size, domain, and optimization choices, measuring held-out accuracy against training FLOPs. With an external post-trained teacher, OPD generally loses to GRPO in accuracy per training FLOP at matched settings, even when the teacher is further trained with GRPO, and often suffers response-length inflation. Yet a different teacher changes the picture: when the teacher is the student's own GRPO checkpoint, OPD recovers that checkpoint's accuracy on math with $50\times$ fewer distillation FLOPs than GRPO needs to reach the same accuracy. Following the same teacher setting, we consistently observe this high efficiency with a larger student, on code, and in a second model family. On a single domain, however, the GRPO checkpoint is already the desired model, so distilling it adds to total compute. Across domains, separate GRPO checkpoints hold different capabilities, and no single checkpoint serves every domain. Multi-teacher OPD (DeepSeek-AI, 2026) consolidates per-domain GRPO experts into a single student that matches each expert's accuracy, with the distillation stage costing only 5.2% (on 2 domains) and 8.0% (on 4 domains) of the specialist training FLOPs. Counting specialist training and distillation together, this pipeline reaches joint GRPO's performance on the 2-domain mix with 42.0% fewer total FLOPs.

AI 总结：本研究开发MAEMs元模型，通过无标签RL目标将语言模型残差流方向映射到强激活文本，其在激活反转、跨域性能等方面优于基线，可作为元模型训练的有依据替代方案。

链接：https://arxiv.org/abs/2611.08465

机构：MATS； Arb Research； Charles University（查理大学）； Anthropic

作者：Celeste De Schamphelaere, Juan J. Vazquez, Ari Spiesberger, Tomáš Gavenčiak, Gavin Leech, Euan Ong, Jiahai Feng

英文摘要：To interpret a direction in a language model's residual stream, we can look for text that strongly activates it. But when such text is found by scanning a corpus, it is limited to what that corpus contains. We develop Maximally-Activating Example Metamodels (MAEMs), language models which map residual stream directions from a subject model to text exemplars that strongly activate those directions. Our objective is to maximise the cosine similarity of a supplied residual stream direction with the activations of the text generated by the MAEM. This label-free RL objective helps to scale MAEMs on diverse sources of residual stream directions. Empirically, we find the resultant metamodel produces informative exemplars: a Qwen3.6-27B MAEM inverts held-out activations better than corpus search or a natural-language autoencoder (NLA), beats corpus search (at 10M tokens) on a majority of held-out languages/domains, and activates 89% of held-out SAE features. It also identifies concepts behind steering vectors about as well as NLAs, recovers the payloads of LoRA backdoors more often than corpus search, and can even verbalise some model workspace inner monologue. Unlike explanations produced by other metamodels (e.g. NLAs), MAEM exemplars can be verified with a forward pass on the subject model, making the MAEM objective a grounded alternative for scaling the training of metamodels.

AI 总结：该研究构建了含3种规模GPT-2的LLM动物园，用流匹配模型生成预训练LLM权重，可泛化到未见过的数据混合，为权重空间学习扩展到LLM领域提供支持。

链接：https://arxiv.org/abs/2611.07103

机构：Khoury College of Computer Science（科里计算机科学学院）； Northeastern University（东北大学）； MLAI Lab（MLAI实验室）； KAIST（韩国科学技术院）； Slater Lab（斯莱特实验室）； University of Oxford（牛津大学）

作者：Scott Biggs, Bedionita Soro, Saumya Gupta, Bruno Andreis, Sung Ju Hwang

英文摘要：A challenge to Weight Space Learning as it expands from task specific and vision models to Large Language Models (LLMs) is the absence of a well curated and analyzed LLM population. To address this, we train such a model zoo for application to Weight Space Generation by systematically varying the data mixture composition between web text, books, mathematics, coding, and French across populations of 51M, 124M, and 355M parameter GPT-2 Models, each tier with count N = 100. We analyze these populations by cosine similarity, functional diversity, Leave-One-Mixture-Out modeling, and variance decomposition. We show that the zoo contains diverse models whose variation is structured by training mixture composition. We embed the training mixture composition to condition a Flow Matching model of complete pretrained LLM weights in Gram-PCA compressed weight space. By demonstrating zero-shot generalization to held out mixtures, measuring Continuous Ranked Probability Score skill, as well as finding minimal changes in downstream test-set perplexity and general capability benchmark scores, we show that our generator generalizes on the simplex of data mixtures to produce models that perform similarly to those in the zoo ex nihilo. Finally, we produce all source code and artifacts to support the continued expansion of Weight Space Learning and Generation into the LLM domain.

AI 总结：本研究提出多模态潜在扩散方法，通过生成稀有金融事件的合成场景，提升CVaR标记的稀有金融窗口分类性能，相关模型表现优于基线。

链接：https://arxiv.org/abs/2611.07282

机构：The Graduate Center, CUNY（纽约市立大学研究生中心）； Hunter College, CUNY（纽约市立大学亨特学院）

作者：Mahdi Arab Loodaricheh, Anita Raja, Indranil SenGupta

英文摘要： Extreme market movements are scarce, heterogeneous, and difficult to learn from imbalanced historical data. We study whether generative augmentation can improve the representation and classification of CVaR-labeled rare financial windows while retaining the complementary structure exposed by different financial representations. Each market window is expressed through three aligned modalities: structured market variables, a spatial image transformation, and a sequence of portfolio returns. Separate variational autoencoders map the modalities to compact latent codes, and a class-conditional diffusion model learns their joint latent distribution. Conditional Value-at-Risk (CVaR) provides a risk-aware definition of rare windows and directs generation toward the minority regime. Across four window lengths on U.S. technology equities, the resulting synthetic samples improve fidelity and support relative to tabular diffusion baselines. An XGBoost classifier trained with real and generated samples attains AUCs of 0.953 - 0.960, exceeding both tabular generators and the corresponding real-only baselines. These results show that multimodal latent generation can provide useful synthetic stress scenarios for data-scarce financial risk models.

AI 总结：本文提出一种两阶段残差学习方法，通过LSTM网络分别预测太阳辐照和误差分布，可显著降低预测标准差与历史值的偏差，更精准捕捉决策所需的预测不确定性。

链接：https://arxiv.org/abs/2611.07444

作者：RR Flanagan, E Kasha, PW Talbot, AS Epiny, MR Deinert

英文摘要：Accurate forecasting of renewable generation requires not only point predictions but also reliable estimates of the uncertainty attached to them, particularly at extended horizons. Machine learning models are increasingly used to predict solar output, but they typically report a single value and capture the associated uncertainty poorly. The standard remedy is to train a model to predict the parameters of an assumed probability distribution directly. Here we develop an alternative two-stage method, motivated by synthetic solar signal generation, in which the forecast error is modeled separately from the expected irradiance. A first long short-term memory (LSTM) network predicts the expected global horizontal irradiance, and a second network models the distribution of the forecast error, which is added back to the mean to give both the forecast and its uncertainty. Using historical weather data from the National Solar Radiation Database and a probability integral transform evaluation, we show that explicitly modeling the error brings the predicted standard deviation markedly closer to the historical value, reducing the deviation between the two by 34.6 percent for a standalone LSTM and by 56.5 percent when the LSTM is coupled to Gaussian process regression, the most reliable of the configurations tested. Because the method models the total forecast error directly, it captures the full predictive uncertainty that matters for decisions, rather than only the model contribution that dropout-based approaches estimate.

AI 总结：该研究提出基于Flux的零样本编解码器FluxC，分析编码路径与重建方法对率-失真-感知的影响，发现ODE重建在低率下失真更低，且FluxC性能优于部分现有神经图像压缩方法。

链接：https://arxiv.org/abs/2611.07929

机构：Hosei University（法政大学）

作者：Fuma Kimishima, Jinjia Zhou

英文摘要：We propose FluxC, a zero-shot codec that uses Flux as a compression prior. Although DiffC has shown that pretrained diffusion models can serve as zero-shot codecs, we ask how the choice of coding path and reconstruction method affects rate--distortion--perception behavior. We analyze FluxC along three axes: coding cost and bit allocation, reconstruction method, and low rate distortion. At matched signal-to-noise ratio (SNR), we show that the path choice does not affect the ideal information rate, but changes how bits are allocated across time steps. For reconstruction, we compare ancestral sampling, one-step ODE reconstruction, and many-step ODE reconstruction. Our analysis shows that ODE-based reconstruction can achieve lower distortion than ancestral sampling, especially in the low rate regime. Experiments on Gaussian sources and natural images support this prediction. Moreover, FluxC performs competitively with, and in some cases outperforms, recent neural image compression methods, suggesting that Flux can serve as a principled zero-shot codec.

AI 总结：针对卵巢癌多组学合成数据评估问题，提出生物学导向的评估框架，结合变分自编码器架构生成配对多组学数据，为多组学生成器比较提供参考。

链接：https://arxiv.org/abs/2611.08394

机构：University of Helsinki（赫尔辛基大学）； Orion Pharma（奥利安制药）； PalsaIQ

作者：Déborah Boyenval, Alina Malyutina, Jaana Oikkonen, Sammeli Liikkanen, Sampsa Hautaniemi

英文摘要：Paired multi-omic cancer cohorts are costly to assemble and difficult to share, motivating synthetic cohorts as a complementary resource. As generative model development accelerates, we shift the focus from proposing yet another architecture to defining evaluation criteria that determine whether synthetic data preserve biologically relevant structure. We introduce a biology-informed and patient-disjoint framework for evaluating the generation of paired RNA expression and copy-number variation (CNV) profiles, and apply it to the longitudinal DECIDER high-grade serous ovarian cancer cohort. The framework assesses modality-specific fidelity and coherence between RNA and CNV pairs using agreement between real training and held-out data as a reference, alongside complementary training-data proximity diagnostics. We adapt three multi-omic integration architectures based on variational autoencoders to generate paired profiles. The model leading several fidelity and coherence measures also shows the strongest training-data proximity. The protocol and case study offer the community a reference for comparing future multi-omic generators.

AI 总结： 本文研究一般和博弈中乐观对冲的后悔特性，发现其后悔界与玩家数量相关，两人博弈中可实现O(n log n)的与时间无关的后悔，三人及以上博弈中则存在指数级或线性增长的下界，还提出了简化的时间近似分析框架。

链接：https://arxiv.org/abs/2611.07078

作者：Yuval Dagan, Nadav Eisen, Liad Erez, Tomer Koren

英文摘要：Optimistic Hedge, a multiplicative weight update with one step of extrapolation, is one of the most natural regret-minimizing dynamics for games, and characterizing its regret has been a longstanding open question. Daskalakis, Fishelson and Golowich proved that if all $m$ players in a finite general-sum game follow Optimistic Hedge with a fixed learning rate, each player's regret is polylogarithmic in the horizon $T$. Subsequently, Ha showed that an appropriate choice of learning rate makes each player's regret bounded independently of $T$. This bound, however, is non-constructive and its dependence on the number of actions $n$ is not explicit, leaving open whether Optimistic Hedge can achieve a constant regret bound which is polynomial in the size of the game. We show that the answer hinges on the number of players. In two-player games, there is a learning rate for which Optimistic Hedge guarantees $O(n\log n)$ regret independent of the horizon. In three-player games, however, no such bound is possible: for any learning rate up to an absolute constant and for sufficiently large horizon $T$, there is a game with $n$ actions per player in which some player's regret is $2^{\Omega(n)}$, and more generally, for every $m\ge3$ we construct $m$-player games in which some player's regret is $2^{\Omega(mn)}$; for learning rates above $2^{-\Omega(mn)}$, that player's regret even grows linearly in $T$. Our upper bound is based on a temporal approximation framework for the analysis of Optimistic Hedge. Starting from the variance-based RVU bound of Daskalakis et al., we approximate the losses over time-localized windows by well-behaved functions, whose total variation is a small fraction of their cumulative variance. In $m$-player games, the same framework gives a significantly simplified proof of the recent $O(\sqrt{m}\log n\log T)$ regret bound due to Ha.

AI 总结：该研究针对两阶段在线优化问题提出Meta-Frieren通用框架，推导元级选择准则与Feel-Good Thompson Sampling的高概率悔界，实验验证其可用于多类别模型的超参数优化。

链接：https://arxiv.org/abs/2611.07169

机构：CREST, ENSAE, IP Paris（CREST、ENSAE、巴黎理工学院）

作者：Vahan Arsenyan, Arnak Dalalyan

英文摘要：We study a two-stage online optimization problem: at each round, a meta-learner first selects an agent-environment pair, then the chosen agent interacts with the environment, and receives a reward. We introduce Meta-Frieren, a general framework for maximizing cumulative reward in this setting. Our analysis yields a simple UCB-style criterion for meta-level selection and recovers prior guarantees as special cases while extending them to broader settings. We instantiate Meta-Frieren with GP-UCB under Bayesian and frequentist assumptions. We also derive, to our knowledge, the first high-probability regret bound for Feel-Good Thompson Sampling, specialize it to finite and parametric function classes, and integrate it into Meta-Frieren. Experiments support our theoretical results and demonstrate that Meta-Frieren can be successfully applied to hyperparameter optimization across multiple model classes.

AI 总结：针对带记忆和时变约束的在线凸优化框架COCO-M研究不足的问题，提出基于延迟等价的乐观算法，利用记忆到延迟的归约，建立了适配环境非平稳性和预测精度的子线性 regret 界，提升了算法性能。

链接：https://arxiv.org/abs/2611.08302

机构：L2S, CentraleSupélec（L2S中央高等电力学院）； Université Paris-Saclay（巴黎萨克雷大学）； CNRS（法国国家科学研究中心）； Delft University of Technology（代尔夫特理工大学）； Télécom SudParis（巴黎高等电信学院）； Institut Polytechnique de Paris（巴黎综合理工学院）

作者：Mohammed Abdullah, George Iosifidis, Salah Eddine Elayoubi, Tijani Chahed

英文摘要：Online Convex Optimization with memory and time-varying constraints (COCO-M) is a recently-introduced framework that captures the dynamics of stateful online learning and non-stochastic control with budget constraints. Despite its expressive power, COCO-M remains largely understudied. In this work, we tackle this problem in its full scale, providing an optimistic algorithm that incorporates untrusted predictions to tame universal dynamic regret and cumulative constraint violation. Our analysis leverages a potent reduction from memory to delay and treats state-dependent gradients as delayed feedback. We establish the first bounds that adapt to both environment non-stationarity (path length $P_T$) and prediction accuracy, while lifting previous restrictive assumptions such as functional separability. En route to these results, we derive findings of independent interest for the memory-to-delay reduction, and for delayed OCO with time-varying constraints. The proposed framework ensures sublinear worst-case guarantees that improve with prediction accuracy, reaching $\mathcal O(1)$ regret under perfect predictions, with cumulative constraint violation (CCV) $\mathcal O(\sqrt{T})$ for unknown $P_T$ and $\mathcal O(\log T)$ for known or online observable $P_T$, while recovering the optimal COCO rates when memory disappears.

AI 总结：提出LowRAM，通过测试时的低秩键压缩与值缓存摊销组合，在近无损性能下实现KV缓存的高效压缩，提升解码吞吐量，且可与量化、驱逐进一步组合以获得更高压缩比。

链接：https://arxiv.org/abs/2611.07025

机构：University College London（伦敦大学学院）

作者：Teresa Delgado, Hugo Hazard, Martin Benfeghoul, Jun Wang, Zafeirios Fountas

英文摘要： Autoregressive decoding relies on the KV cache, whose memory footprint grows linearly with context length and caps batch size and throughput. Existing post-training methods exploit redundancy across tokens, layers, and feature dimensions, or lower numerical precision. We identify a complementary axis: within each attention head, values are largely predictable from their corresponding keys. We exploit this with value cache amortisation, which fits a lightweight per-head predictor at test time and stores residuals only for the tokens it reconstructs worst, giving $4\times$ near-lossless value compression. The predictor can be initialised by meta-learning for substantially reduced prefill overhead, or fitted from scratch in settings that allow a one-off cost, such as multi-query inference. We pair value amortisation with cross-layer low-rank key compression to produce LowRAM, an asymmetric compression pipeline that stays close to lossless across three model families. By reconstructing only a selected subset of KV pairs at each decoding step, LowRAM can keep decoding overhead minimal and achieve up to $5\times$ higher peak decode throughput at $4\times$ compression. LowRAM further composes with quantisation and eviction, reaching $26.9\times$ compression at $94.5\%$ of baseline performance on long contexts. These results show that treating keys and values as structurally distinct enables efficient, aggressive, and composable KV cache compression.

AI 总结：提出DISCERN方法，用类参数分布采样的合成嵌入替代保留集，在Transformer等骨干的增量类遗忘任务中，能稳定减少遗忘类重新出现，表现优于多数基线。

链接：https://arxiv.org/abs/2611.07892

机构：University of Granada（格拉纳达大学）； Andalusian Institute of Data Science and Computational Intelligence (DaSCI)（安达卢西亚数据科学与计算智能研究所）； Department of Computer Science and Artificial Intelligence (DECSAI)（计算机科学与人工智能系）

作者：Nicolás Segura-Kunsagi, Juan Luis Suárez-Díaz, Iván Sevillano-García, Francisco Herrera

英文摘要：Machine unlearning aims to remove the influence of specific training data from a trained model without retraining from scratch. In real deployments, however, deletion requests arrive over time, calling for incremental class unlearning (ICU): classes are removed sequentially, previously forgotten knowledge must remain forgotten, and the remaining classes must be preserved. We present DIstributional SynthetiC rEplay for Retain-free iNcremental unlearning (DISCERN), an intervention-based class unlearning method that replaces access to the retain set with synthetic embeddings sampled from per-class parametric distributions. Unlike prototype-based replay, which assumes an isotropic Gaussian around each class centroid, DISCERN fits Euclidean (Gaussian, Laplace, Student's $t$) and directional (von Mises--Fisher, Fisher--Bingham) families to each class and selects among them through statistical tests and, when none of them accepts, cross-validated log-likelihood. This design is motivated by the architecture-dependent latent geometry: transformer embeddings lie approximately on a hypersphere and concentrate around class directions, unlike those of convolutional networks. We evaluate DISCERN against seven retain-free baselines on MNIST, CIFAR-10, and CIFAR-100 with two convolutional and two transformer backbones, under single-class, multi-class, and incremental unlearning. DISCERN is competitive when classes are removed at once and is the most stable method in ICU on transformers: on ViT-B/16, after 20 sequential deletions it retains a harmonic mean of $0.797$, whereas the strongest method in the simultaneous setting drops from $0.918$ to $0.048$. A paired ablation shows that per-class distribution selection substantially reduces the resurfacing of previously forgotten classes, acting as a stabilizer of accumulated forgetting rather than a uniform improvement.

AI 总结：针对医疗场景中测试数据分布不可知的长尾联邦学习问题，提出GTAFL框架，通过训练阶段的自适应重采样等技术和推理阶段的集成机制，在CIFAR10及两个医学数据集上性能优于现有方法。

链接：https://arxiv.org/abs/2611.08021

机构：School of Computer Science and Engineering, Huizhou University（惠州学院计算机科学与工程学院）； College of Engineering, Northeastern University（东北大学工程学院）； College of Computer Science, Beijing Institute of Technology, Zhuhai（北京理工大学珠海学院计算机学院）

作者：Guangyu Chen, Jingyun Zeng, Hui Jiang

英文摘要：In healthcare, protecting patient privacy is crucial due to the sensitivity of medical data and its extensive accessibility. Federated Learning (FL) offers a decentralized and privacy-preserving training paradigm, making it an ideal solution for healthcare applications. A critical challenge in healthcare is that real-world medical data often exhibits long-tailed distributions in local and global views. Existing methods addressing long-tailed FL problem typically assume that the model will be evaluated on uniform test data distribution. However, practical test data in healthcare systems is often agnostic and unpredictable, leading to potential model failures in real-world scenarios. In this paper, we introduce a novel task termed Test-Agnostic Long-Tailed Federated Learning and propose GTAFL, a comprehensive framework to address this challenge. During the training stage, GTAFL employs adaptive re-sampling, expert classifier retraining, and self-supervised learning to correct biased classifiers and distorted feature spaces caused by long-tailed training distributions. During the inference stage, an ensemble mechanism combines retrained expert classifiers to handle test data with unknown distributions. Extensive experiments on CIFAR10 and two medical datasets manifest that our framework outperforms other state-of-the-art methods. The code is available at this https URL

AI 总结：该研究针对贝叶斯M/EEG源成像的不确定性校准问题，提出模拟框架，对比BMN+sLORETA与SBL+S-FLEX方法，发现等渗校准可提升性能，SBL+S-FLEX在定位和不确定性上更优。

链接：https://arxiv.org/abs/2611.06979

机构：Technische Universität Berlin（柏林工业大学）； University of California San Francisco（加州大学旧金山分校）； Charité-Universitätsmedizin Berlin（柏林夏里特医学院）

作者：Ismail Huseynov, Mohammad Orabe, Srikantan Nagarajan, Stefan Haufe

英文摘要： M/EEG source imaging requires accurate localization and reliable uncertainty estimates, yet posterior credible regions are rarely assessed for empirical calibration. We present a simulation-based framework for uncertainty quantification and calibration of empirical Bayesian inverse methods with closed-form conditional Gaussian posterior summaries. We compare Bayesian minimum norm (BMN) with sLORETA depth weighting and sparse Bayesian learning (SBL) with sparse basis field expansions (S-FLEX). S-FLEX allows retained coefficients to contribute nonzero marginal posterior variance at locations where voxel-wise SBL yields degenerate variances. For BMN, we derive joint source- and noise-variance updates through evidence maximization. We test whether adaptive noise learning yields near-calibrated uncertainties. Both methods retain residual miscalibration, motivating post-hoc isotonic recalibration of credible-region levels using empirical coverage. Simulations assess fixed- and free-orientation models across signal-to-noise ratios (SNR), source sparsity levels, and subject-specific head models, using oracle, baseline-estimated, and adaptively learned noise. Before recalibration, SBL+S-FLEX yields lower Wasserstein localization error, smaller posterior uncertainty, and coverage closer to nominal credibility than BMN+sLORETA. Isotonic recalibration improves both methods, with larger gains for conservative BMN+sLORETA posteriors; transfer across mismatched SNR, sparsity, and head models remains condition dependent. Noise specifications yield broadly consistent reconstruction and uncertainty trends within each estimator. In real auditory MEG data, SBL+S-FLEX provides more focal localization and tighter uncertainties than BMN+sLORETA, suggesting greater statistical power for identifying bilateral auditory sources.

AI 总结：本研究收集291项熔盐氯化物腐蚀试验数据，引入CSPI指数，采用混合树集成模型开发可解释机器学习框架，实现可靠腐蚀速率预测，为熔盐腐蚀建模提供新方法并指出改进方向。

链接：https://arxiv.org/abs/2611.07125

机构：North Carolina State University（北卡罗来纳州立大学）

作者：Md Mahmudul Hasan Pathik

英文摘要：Molten chloride corrosion has been studied for decades, yet reliable prediction across different alloys, salts, and test conditions remains difficult. In this work, 291 static molten chloride corrosion experiments reported in the literature were compiled into a structured dataset and used to develop an interpretable machine-learning framework for corrosion-rate predictions. To better represent the differences in salt condition across studies, a Chloride Salt Preparation Index (CSPI) was introduced to bring the reported salt-treatment method, atmosphere control, and pre-exposure impurity condition on a common numerical scale. A blended tree-based ensemble provided the most consistent predictive performance, achieving a cross-validated \(R^2\) of 0.85 and an MAE of 0.41 mg cm\(^{-2}\) day\(^{-1}\). The graded CSPI retained more useful information than a simple purified/unpurified classification and emerged as the dominant model descriptor. Global sensitivity and SHAP analyses further identified exposure time, temperature, and alloy chemistry as important contributors to the model predictions, with several observed trends consistent with the established molten-chloride corrosion behavior. The conformal prediction intervals were well calibrated, providing an estimate of prediction uncertainty in addition to the point prediction. Overall, the framework demonstrates that historical corrosion data can support quantitative, interpretable, and uncertainty-aware prediction when experimental context is represented explicitly, while also highlighting where more consistent reporting and targeted experiments are needed to improve future molten-salt corrosion models.

AI 总结：该研究提出基于Phikon嵌入的可扩展贝叶斯GPC框架，结合RPCholesky、HMC等技术，在PCam和CAMELYON17-WILDS数据集上，较基线大幅降低校准误差与假阴性率，提升分布偏移下的不确定性可信度。

链接：https://arxiv.org/abs/2611.07127

机构：New York University（纽约大学）

作者：Yunru Zheng, Wun Ting Chan, Sampras Dsouza, Zhenyu Hong

英文摘要：Reliable uncertainty quantification is essential for clinical deployment of cancer diagnosis models, yet neural networks and variational Gaussian process methods can produce overconfident predictions, particularly under the distribution shifts that arise when models trained at one hospital are applied at another. We present a scalable Bayesian Gaussian Process Classification (GPC) framework that operates on frozen pathology foundation model embeddings (Phikon) and combines accelerated Randomly Pivoted Cholesky (RPCholesky) for low-rank kernel factorization, Hamiltonian Monte Carlo (HMC) with non-centered parameterization for posterior sampling, and a pivots-as-inducing-points scheme for prediction. This makes HMC posterior sampling in the low-rank latent representation tractable at N approximately 300,000 training points, enabling HC-based low-rank GPC on substantially larger datasets than prior MCMC-based demonstrations. We evaluate on two histopathology benchmarks for breast cancer metastasis detection: PCam and CAMELYON17-WILDS, where the latter tests calibration under hospital-level distribution shift. Compared to stochastic variational GP and neural network baselines, our method improves calibration and predictive performance - reducing expected calibration error by up to 88% and false negative rate by up to 90% - with the largest gains appearing under distribution shift, suggesting that HMC-based posterior sampling provides more trustworthy uncertainty estimates for cross-site pathology classification.

AI 总结：本文提出轻量级事后方法TrustMix，无需重训预训练模型，通过结合模型预测分布与用户指定的 fallback 分布实现分布偏移下的优雅降级，在多任务多数据集上验证其可提升预测不确定性与决策可用性。

链接：https://arxiv.org/abs/2611.07206

机构：University of Oxford（牛津大学）

作者：Juliusz Ziomek, Maike Osborne, Yee Whye Teh

英文摘要：Predicting beyond the training domain necessarily requires assumptions about how the ground truth extrapolates, yet in most neural networks these assumptions are made only implicit through the architecture. We propose TrustMix, a lightweight post-hoc method that makes a pretrained model's extrapolative behaviour explicit and controllable. TrustMix combines the predictive distribution produced by a pretrained model with a user-specified fallback distribution $p_0(y|x)$, such that as the out-of-distribution (OOD) deviation statistic grows, we collapse towards $p_0(y|x)$. TrustMix provides graceful degradation: the predictive distribution converges toward the fallback distribution as the OOD signal grows. The fallback distribution $p_0(y|x)$ can reflect domain knowledge, if available, or rely on a domain-agnostic stationarity assumption. TrustMix does not require retraining of the base model and can be applied to both regression and classification problems. We evaluated TrustMix on several synthetic and real-world distribution shifts, including tabular, image, and natural language processing tasks. We tested TrustMix on datasets such as CIFAR-10/100 and ImageNet and on numerous architectures including Convolutional Neural Networks (CNNs) and Transformers. TrustMix remains competitive in-distribution while improving predictive uncertainty under distribution shift, with gains in predictive likelihood for regression and calibration for classification. Finally, we showed that TrustMix can provide useful predictive distributions for decision-making applications by utilizing it as a surrogate for Bayesian Optimisation.

AI 总结：TriAttention通过将QK分块分配给跳过、4位或8位计算，结合压缩KV缓存，在保持模型质量的同时，使长上下文LLM推理速度最高提升7.88倍。

链接：https://arxiv.org/abs/2611.07256

机构：University of North Texas（北得克萨斯大学）； Saint Louis University（圣路易斯大学）

作者：Hanzhi Zhang, Qiao Zhang, Qinglei Cao, Heng Fan, Yan Huang, Kewei Sha, Yunhe Feng

英文摘要：Sparsity and quantization reduce the cost of attention by skipping interactions and lowering the precision of those retained. However, useful interactions differ in their precision requirements: some tolerate low-precision scores while remaining sensitive to removal. We therefore introduce TriAttention, which assigns query-key (QK) tiles to skipping, 4-bit, or 8-bit computation, using 4-bit to retain quantization-tolerant interactions at lower cost. To represent compactly, we encode each head's computation regions with two diagonal-distance boundaries, keeping policy storage independent of context length. Offline calibration then evaluates candidate boundary pairs using normalized projected-output errors and complete-tile execution costs. Using these measurements, a model-wide allocator selects each head's boundaries to minimize calibration error under a shared computation budget. During prefill, the selected boundaries generate precision-grouped traversals for a fused kernel that skips omitted tiles and combines contributions from both QK precisions into one normalized attention output. The quantized keys and values prepared during prefill are then reused in a compressed key-value (KV) cache for decoding, reducing repeated reads of prompt data. Experiments show that TriAttention achieves end-to-end generation speedups of up to 7.88x over the Torch attention and 2.65x over the BF16 baseline. Meanwhile, TriAttention maintains quality comparable to the original models on general-purpose and long-context tasks. These results show how a compact retention-and-precision policy translates local sensitivities into attention computation and repeated KV-cache access. The implementation is this https URL.

AI 总结：该研究探究Transformer能否在上下文内控制未知参数的不稳定非线性动力系统，采用GPT-2架构，经特定训练技术后可实现该控制，且性能随上下文提升，还能泛化到分布外系统。

链接：https://arxiv.org/abs/2611.07291

机构：University of California Berkeley（加州大学伯克利分校）； University of Illinois, Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； University of Southern California（南加州大学）； University of California Merced（加州大学默塞德分校）

作者：Ebonye Smith, Aidan Andrews, Phoenix Wilson, Alexis Frias, Ayush Pandey, Gireeja Ranade

英文摘要：One of the remarkable abilities of Transformer-based neural networks is in-context learning. This motivates us to think about "in-context" control. Can we control a system knowing only (1) that it belongs to a particular family of systems and (2) some data on system evolution (provided as the context trajectory) but no specific parameter information? Specifically, we investigate the ability of Transformer-models to perform "in-context" control of unstable, nonlinear dynamical systems, where context consists only of the unforced system dynamics. We leverage a multi-head, GPT-2-style architecture, and we evaluate performance for two classical nonlinear systems -- the cartpole system and the two-link acrobot system, where the system parameters are unknown to the Transformer. We consider a setting where trajectories from an expert controller are available at training time but not at inference time. Here, we identify training techniques that are essential to train Transformers to perform low-level control. Through an analysis over checkpoints, we show that in-context control is possible, and the learned policy performance generally improves as context increases. Furthermore, the Transformer models can generalize to stabilize systems that are out-of-distribution.

AI 总结：本文通过白盒与黑盒攻击，实证分析核机器等模型的对抗鲁棒性，提出核机器近似对抗训练算法，发现核机器固有鲁棒性优于神经网络。

链接：https://arxiv.org/abs/2611.07474

机构：Center for Machine Intelligence and Data Science, IIT Bombay（印度理工学院孟买分校机器智能与数据科学中心）

作者：Rahul Kumar Yadav, Tunir Ghosh, Sudhendu Ahir, Parthe Pandit, Arjun Bhagoji

英文摘要：Kernel machines are an important class of models in machine learning. However, the question of how robust modern kernel machines are to adversarial examples has not been studied comprehensively. In this paper, we provide a thorough empirical investigation of the adversarial robustness of kernel-based models for both binary and multiclass classification tasks, via white-box and black-box evasion attacks. We evaluate the native robustness of kernel machines, random feature classifiers, and the relatively newer Recursive Feature Machines. In addition, we provide an algorithm for approximate adversarial training of kernel machines. Our experiments on classification tasks on image and tabular datasets strongly indicate that kernel machines are natively more robust than neural networks, i.e., their performance degrades more gracefully against adversarial perturbations.

AI 总结：该研究针对神经算子训练需昂贵输入-解对的问题，提出WaSP流水线，通过分阶段合成数据减轻分布偏移，使PDE解误差降低3.9至9.5倍，且能更好重现混沌系统统计可观量。

链接：https://arxiv.org/abs/2611.07646

机构：California Institute of Technology（加州理工学院）； University of Toronto（多伦多大学）

作者：Ryan Liu, Chuwei Wang, Ricardo Baptista, Andrew Stuart, Anima Anandkumar

英文摘要： Neural operators offer fast surrogates for solving PDEs, but training them typically requires expensive input-solution pairs. Two solver-free alternatives include physics-informed learning and training on manufactured data. Physics-informed learning approximates solutions by minimizing PDE residuals. While the residual norm must be strong enough to control solution error, stronger norms increase mini-batch variance and hinder optimization. Manufactured data constructs input-solution pairs for learning by applying the PDE operator to artificial solutions. However, the resulting training distribution can be mutually singular with the target distribution, yielding poor accuracy on the target distribution. We introduce the Wasserstein Data Synthesis Pipeline (WaSP), which trains on manufactured input-solution pairs while gradually mitigating the distribution shift. WaSP synthesizes new pairs by applying the learned operator to a path of intermediate distributions converging to the target input distribution. Training proceeds in stages along this path. At each stage, the previous surrogate generates solutions, and the exact PDE operator derives corresponding inputs, yielding training pairs for the next surrogate. Theoretically, we prove a stepwise error-propagation bound for the distribution shift introduced during synthesis. Empirically, WaSP reduces solution error by $3.9\times$ to $9.5\times$ compared to existing solver-free methods on PDEs such as Allen-Cahn, Kuramoto-Sivashinsky, and Ginzburg-Landau, and better reproduces the two chaotic systems' statistical observables.

AI 总结：该研究针对表格基础模型的主动上下文学习，对比TANN、EPIC等方法与经典规则，发现EPIC在标签预算下表现最优，仅用少量标签即可保留大部分全数据性能。

链接：https://arxiv.org/abs/2611.07886

机构：University of Technology Nuremberg（纽伦堡工业大学）

作者：Christian M. M. Frey

英文摘要：Tabular Foundation Models (TFMs) adapt to a new task only through the labeled rows in their context. However, labels are expensive to acquire in real-world scenarios due to the need for highly specific domain knowledge to ensure high-quality data. Active in-context learning transfers classical acquisition rules to this setting, in which the model that guides acquisition changes with every label added. We ask how the value of diversity, predictive uncertainty and target-aware information gain changes as a frozen TFM acquires more context. For that, we study Target-Aware Nearest Neighbor (TANN) acquisition and the Expected Predictive Information Gain in Context (EPIC) and evaluate them against classical rules. Our study across twenty benchmarks with two well-established TFMs and classical GBDTs highlights three findings: (i) When: diversity-based acquisition is best during the first acquisition iterations; (ii) Where: labeling pool rows that serve the target inputs via a simple target-aware nearest-neighbor rule matters substantially; (iii) How: EPIC yields the best models at the label budget, retains on average 80--87% of the chance-corrected full-data performance with only 5-10% of the pool labeled on the eight benchmarks with the largest pools, and selects labels that also improve the predictive performance of gradient-boosted decision trees.

AI 总结：本文研究大步长对抗训练中梯度下降的动力学，明确其与标准训练的三个关键差异，加深了对大步长对抗训练行为的理解。

链接：https://arxiv.org/abs/2611.08457

机构：Aarhus University（奥胡斯大学）

作者：Yi Feng, Andrea Paudice

英文摘要：Adversarial Training (AT) is an important method for improving model robustness. We study the dynamics of deterministic Gradient Descent (GD) in AT. Most existing works in this direction focus on the small or vanishing step sizes regime, where GD resembles a continuous-time flow. However, such step sizes can make training inefficient and are rarely used in practice. In this paper, we study the AT dynamics of GD with large constant step sizes. We first study GD in adversarial linear logistic regression, focusing on two aspects: where the model parameters converge (implicit bias) and how quickly the losses decrease (optimization). We then turn to neural networks training and study how the sharpness of losses evolve during AT. Our results advance the understanding of the behaviors of GD with large step sizes in AT and highlight three key differences from its behavior in Standard Training (ST): (i) the implicit bias gap between AT and ST, previously established only for small step sizes, persists for arbitrarily large constant step sizes, (ii) different from ST, there is no uniform rate of convergence for the optimization error in AT, and (iii) sharpness can exhibit two-stage dynamics that differ from the Edge-of-Stability behavior observed in ST.

AI 总结：针对车载ISAC网络中功率分配的非凸优化问题，提出GNN-ISAC算法，其性能接近WMMSE且计算量小，泛化能力优于异构GNN和信道盲基线。

链接：https://arxiv.org/abs/2611.07277

机构：University of Melbourne（墨尔本大学）； Monash University（莫纳什大学）

作者：Junyu Liao, Lili Chen, Tian Han, Jamie Evans

英文摘要：In vehicular integrated sensing and communication (ISAC) networks, each service-provider vehicle uses a limited power budget to serve multiple sensing targets and communication receivers. Maximizing the communication sum-rate subject to sensing signal-to-interference-plus-noise ratio (SINR) constraints is nonconvex, and iterative optimization is costly to rerun for every network layout. This paper proposes a graph neural network (GNN)-based power allocation algorithm, termed GNN-ISAC. Concretely, GNN-ISAC addresses the aforementioned problem by considering two distinct edge types in the graph representation, one capturing the inter-provider interference, the other representing the intra-provider power-budget sharing. GNN-ISAC is trained without labels using a sensing-penalized sum-rate loss. We also adapt the weighted mean-square error minimization (WMMSE) algorithm to this problem by introducing a sensing-power reservation step. Simulation results show that GNN-ISAC achieves communication sum-rates close to WMMSE with low sensing-link violation rates and much less computation. It also outperforms an existing heterogeneous GNN and a channel-blind baseline, and generalizes to larger networks without retraining.

AI 总结：该研究提出光谱对称指纹（SymFP），利用等变表示的奇异值光谱表征对称性，可实现点群识别、对称性分析，在多类物理系统中表现优于全局几何度量。

链接：https://arxiv.org/abs/2611.07410

机构：Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）； Beijing Academy of Artificial Intelligence（北京人工智能研究院）

作者：Jiacheng Cen, Anyi Li, Ning Lin, Zihe Wang, Ji-Rong Wen, Wenbing Huang

英文摘要：Symmetry governs the organization of physical structures, while symmetry breaking reveals their transformations. In geometric deep learning, symmetry imposes rank constraints on equivariant representations, limiting the expressive capacity of equivariant graph neural networks (GNNs). Here we show that these constraints can instead be exploited: the singular-value spectra of equivariant representations provide a unified fingerprint of structural symmetry. We introduce spectral symmetry fingerprints (SymFP), which enable discrete point-group identification through rank-template matching and continuous symmetry analysis by quantifying spectral energy beyond target rank constraints. SymFP is orientation invariant and architecture agnostic, enabling a parameter-free GNN implementation. Its spectra can be reused across multiple target groups, and the resulting leakage scores satisfy subgroup monotonicity. Across synthetic structures, molecules, coordination complexes and large biological assemblies, SymFP accurately identifies point groups, tracks controlled symmetry breaking and reveals local and degree-resolved symmetry changes that can be obscured by global geometric measures. These results establish the spectral structure of equivariant representations as a general framework for characterizing symmetry across physical systems.

AI 总结：提出GraphMuon图感知优化器，适配Muon矩阵更新与图传播输入协方差结构，在3个OGB基准上较AdamW获1.32-2.42倍加速，15种配置中13种目标时间更短且性能更优。

链接：https://arxiv.org/abs/2611.08379

作者：Guy Hadad, Haggai Roitman, Moshe Eliasof, Bracha Shapira

英文摘要：Different optimizers impose different geometries on parameter updates. Adaptive methods such as Adam and AdamW use coordinate-wise updates, while matrix-aware optimizers such as Muon utilize weight-matrix structure through orthogonalized updates. For Graph Neural Networks (GNNs), however, the effective geometry of its weight matrices is shaped not only by its learned channel-mixing weights, but also by the graph-propagated representations it acts on. Thus, even if an update is well normalized as a matrix, its effect on the layer output can be amplified when applied to graph-propagated features. To address this gap, we introduce GraphMuon, a graph-aware optimizer that adapts Muon's matrix updates to the covariance structure of graph-propagated layer inputs. GraphMuon performs orthogonalization in a metric defined by these propagated representations and normalizes step sizes according to the update's induced effect on the layer representation. We show that theoretically, vanilla Muon can suffer from graph-induced update amplification, while our GraphMuon can mitigate this effect. As such, our GraphMuon yields a graph-metric analogue of Muon's polar update step. Empirically, GraphMuon achieves $1.32\times$-$2.42\times$ validation-loss time-to-target speedups over AdamW on three large-scale OGB benchmarks. Across 15 dataset-architecture configurations spanning node classification, inductive graph classification, and graph regression, it achieves lower time-to-target in 13 cases, while maintaining competitive or superior predictive performance. Code is available at this https URL.

AI 总结：本文提出一种基于历史性能的元学习表示，结合线性贝叶斯岭回归模型，在TabArena基准的HPO与CASH任务上，以更少试验次数达到相当性能，且复杂度低、无需预训练。

链接：https://arxiv.org/abs/2611.07659

机构：Delft University of Technology（代尔夫特理工大学）

作者：Emiel Witting, Cheng Yan, Tom Julian Viering

英文摘要：Bayesian optimization for Hyperparameter Optimization (HPO) and Combined Algorithm Selection and Hyperparameter Optimization (CASH) often relies on explicit hyperparameter encodings as surrogate inputs. In high-dimensional search spaces, however, such representations can make it difficult for standard surrogates to learn useful performance models. We instead study an alternative meta-learning representation that describes each configuration directly as its historical performance across prior tasks. This representation can be paired directly with standard surrogate models. Our scope is the iterative selection component of AutoML, deciding which configuration to evaluate next from a fixed candidate set, rather than a complete framework with features such as runtime estimation, cost-aware budgeting and ensembling. We further apply this representation to maximal marginal relevance (MMR), a diversity-based re-ranking criterion from recommender systems, as a posterior-free acquisition rule. When selecting and optimising across multiple models on the TabArena benchmark, the performance-based representation paired with a linear Bayesian ridge regression model outperforms dedicated meta-learning baselines, achieving the same tuned performance in roughly half the number of trials. For single-model HPO, it remains a competitive second. Across both benchmarks this method closely matches or improves on regret at a fraction of the complexity and cost, requiring no pre-training or tuning and scaling independently of search space dimensionality. Code for reproducing our results is available at this https URL.

AI 总结：本文将LoRA适配器视为数据模态，训练Transformer自编码器构建其非线性潜在空间，发现该空间鲁棒性更高，还可扩展以容纳混合秩适配器，为泛化适配器表示提供了路径。

链接：https://arxiv.org/abs/2611.07775

机构：University of St.Gallen（圣加仑大学）； Université de Montréal（蒙特利尔大学）； Weight Space Labs（权重空间实验室）

作者：Moritz Grävinghoff, Konstantin Wehmeyer, Aron Asefaw, Boris Knyazev, Damian Borth

英文摘要： LoRA adapters are usually treated as finished artifacts rather than as a data modality. Prior work has shown that LoRA adapters populate a low-dimensional manifold that a linear PCA basis already organizes well enough to support different operations such as sampling, editing, and inversion. We ask what a learned, nonlinear latent adds beyond this. We train a transformer autoencoder on the same population of LoRA identity adapters, mapping each into a latent space and compare every operation against the identical operation applied directly to the raw adapter weights. The latent preserves the capabilities already available in weight space, sampling diverse identities and editing individual attributes. Beyond preservation, we show that the latent is far more robust to displacement than weight space, degrading into distorted but still valid faces rather than texture. Fixed-dimensional weight-space representations cannot natively host adapters of more than one rank, so we extend our sequential autoencoder to hold mixed-rank adapters within one shared latent. This gives a path toward treating adapter weights of varying architecture as a single, general modality rather than one tied to a fixed rank, and points to a learned, nonlinear latent as the mechanism that could carry this generalization further, toward heterogeneous populations of adapters.

AI 总结：本研究系统评价IVF/ICSI术后活产预测的机器学习模型，荟萃分析其性能，发现未明确决策点的模型AUC最高，模型提升源于分期预测与局部校准，而非深层架构，且多数研究偏倚风险高。

链接：https://arxiv.org/abs/2611.07124

机构：School of Computer Science, Wuhan University（武汉大学计算机学院）； Reproductive Medical Center, Zhongnan Hospital of Wuhan University（武汉大学中南医院生殖医学中心）； Clinical Medicine Research Center of Prenatal Diagnosis and Birth Health in Hubei Province（湖北省产前诊断与出生健康临床医学研究中心）

作者：Abdul Raheem Shahzad, Muhammad Zaheer Khan, Umar Ajaib, Asad Ur Rehman, Juan Deng, Ming Zhang

英文摘要：Machine-learning models predicting live birth after in vitro fertilisation (IVF) or intracytoplasmic sperm injection (ICSI) from routine clinical data report performance from near chance to near perfect. We systematically reviewed these models under PRISMA 2020, searching OpenAlex, Scopus and Web of Science to 22 October 2026, appraised them with PROBAST+AI and TRIPOD+AI, and pooled their discrimination with random-effects meta-analysis on the logit scale. We define what fixes the meaning of a model's output (decision point, endpoint, unit of analysis, data source) and map the 48 included studies by model family, decision point and validation design. Models evaluated only internally pooled to an area under the receiver operating characteristic curve (AUC) of 0.814, against 0.756 (95% prediction interval 0.546-0.889) for models evaluated on a later period or another centre, and 0.764 for studies not at high risk of bias; neither difference was significant, and heterogeneity was extreme. Models whose decision point was not stated pooled to 0.910, the only significant moderator in meta-regression. Against logistic regression fitted to the same data, the best machine-learning model gained a pooled 0.056 AUC (95% CI 0.017-0.090). Of 48 studies, 47 were at high or unclear risk of bias, only four assessed the calibration slope, and no model has evidence of clinical impact. The certainty of the pooled evidence is low to very low; the gains that survive validation came from staging predictions to decision points and local calibration, not deeper architectures. We provide a TRIPOD+AI-mapped checklist, a benchmark specification and a research agenda.

AI 总结：针对弱依赖等场景，提出QRCIT分位数分辨率条件独立性检验，引入SDB基准，在合成与真实数据集上提升了结构化条件依赖的检测与诊断能力。

链接：https://arxiv.org/abs/2611.07165

机构：Deakin Applied Artificial Intelligence Initiative（迪肯应用人工智能倡议）

作者：Khoa Nguyen, Bao Duong, Thin Nguyen

英文摘要：Conditional independence testing is difficult when dependence is weak, localized, or expressed through distributional changes rather than a global mean shift. Existing tests usually return a scalar statistic or p-value, which summarizes evidence against independence but does not describe where the conditional law changes. We propose QRCIT, a quantile-resolution conditional independence test that compares P(X|Z) and P(X|Y,Z) through a quantile deformation learned by a shared model. A lightweight residual quantile-flow model first learns the baseline conditional quantile law and then learns the additional deformation induced by Y, yielding a global predictive-gain statistic with conditional-randomization calibration and an inspectable deformation field over context and quantile level. We also introduce the Structural Dependency Benchmark (SDB), which evaluates dependence structures motivated by real applications, such as context-local, tail-only, sign-reversing, moving-ridge, and mean-neutral patterns. Across synthetic post-nonlinear (PNL) and SDB benchmarks, QRCIT improves detection of structured alternatives, with empirical Type-I behavior evaluated across settings. Higher quantile resolution improves merged-SDB power at moderate sample sizes, while diagnostics restricted to known affected regions improve detection of localized dependence. On Sachs, METABRIC breast cancer, and CCLE/PLX4720, QRCIT provides competitive real-data performance and interpretable deformation patterns. These results support quantile-resolved conditional-law deformation as a practical framework for testing and diagnosing structured conditional dependence.

AI 总结：该研究揭示重加权虽能改善部分群体间公平性，但会对少数群体引入低估危害，且其效果受数据集特征影响，从业者需结合目标批判性评估该方法。

链接：https://arxiv.org/abs/2611.07258

机构：Cedars-Sinai Medical Center（西达赛奈医疗中心）

作者：Amalia Ionescu, Jose G. Hernandez, Emily F. Wong, Anil K. Saini, Jason H. Moore, Tiffani J. Bright

英文摘要： Reweighting is a widely used pre-processing method for promoting discrimination-free model behavior by assigning different weights to data points during model training. While it is often credited with improving fairness, particularly by mitigating systematic underestimation for underrepresented groups, it can also introduce unintended harms under certain data conditions. Using controlled simulations that vary group size and baseline event prevalence, we show how reweighting can yield misleading improvements in fairness by reducing underestimation for one group while introducing it for another. Across 3,000 simulated datasets with event prevalence of 7-14%, reweighting improved parity in positive classification and false negative rates (FNR) between groups, but did so partly by decreasing positive classifications and increasing FNR for the group with higher baseline prevalence. These effects were most pronounced when that group was a minority: its FNR rose by 6 percentage points, approximately 82 additional missed cases per 10,000 group members, and roughly three-quarters of the improvement in FNR parity came from this increase rather than from gains for the other group. The pattern held across six classifiers, from logistic regression to neural networks, while overall accuracy was essentially unchanged, leaving the shift invisible in aggregate performance metrics. These findings mirror patterns reported in real-world applications and underscore the critical role of data characteristics in shaping the impact of this fairness intervention. While reweighting may create desirable outcomes in some applications, practitioners should critically evaluate it relative to their problem area and goals. Specifically, when the goal is to reduce underestimation in one group without introducing underestimation in another, reweighting may not be the appropriate approach.

AI 总结：本文针对蛋白质基序支架基准的范围与成功标准问题，引入BILBY-PRO基准与BILBY-MD评估框架，发现静态恢复无法保证支架动力学保真度，凸显了三类互补评估维度。

链接：https://arxiv.org/abs/2611.07723

机构：Jülich Supercomputing Centre, Forschungszentrum Jülich（于利希超级计算中心，于利希研究中心）； Center for Advanced Systems Understanding (CASUS)（高级系统理解中心）； Helmholtz Munich（亥姆霍兹慕尼黑中心）； Scientific Computing Center, Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院科学计算中心）； Helmholtz-Zentrum Dresden-Rossendorf e. V. (HZDR)（德累斯顿-罗森多夫亥姆霍兹中心）； Institute of Computer Science, University of Wrocław（弗罗茨瓦夫大学计算机科学学院）； Cluster of Excellence Physics of Life, TU Dresden（德累斯顿工业大学生命物理学卓越集群）； Department of Physics, University of Cologne（科隆大学物理系）

作者：Emile de Bruyn, Moritz Schäffler, Jannik Schneider, Sebastian R. Schmidt, Chanwoong Hwang, Aleena Siji, Utkarsh Upadhyay, Stefan Bauer, Alexander Schug, Stefan Kesselheim, Artur Yakimovich, Alina Bazarova

英文摘要：Protein motif scaffolding benchmarks assess whether generative models can build scaffolds around functionally important residues. Their conclusions depend on which motifs are tested and which criteria define success. Existing benchmarks typically use small hand-curated motif sets and static refolding criteria, but protein function often depends on conformational ensembles and transitions. Here, we address both benchmark scope and success criteria. We introduce BILBY-PRO, a systematically derived benchmark of structurally conserved functional motifs from PROSITE-linked experimental structures, comprising 220 cases from 174 PROSITE motif-pattern entries. Evaluating six recent scaffold-generation models reveals distinct patterns in target coverage and unique-solution yield: RFDiffusion and Proteina solve more targets, whereas ESM3 produces more unique successful scaffolds on the targets it solves. We further introduce BILBY-MD, a reference-based all-atom molecular dynamics evaluation framework. Applied to statically solved designs from the original MotifBench benchmark, it compares motif drift, scaffold drift, motif-contact preservation, and motif-dihedral free-energy landscapes against reference simulations. Of 55 statically solved model-problem pairs, 31 satisfy our MD-supported criterion and only 8 pass all checks. Compact, self-contained motifs are often MD-supported, whereas interface-like or context-dependent motifs fail more frequently. Static refolding success therefore provides useful evidence of local motif recovery but does not establish dynamical fidelity of the full scaffold. BILBY highlights three complementary axes of motif-scaffolding evaluation: target coverage, unique-solution yield, and MD-supported dynamical fidelity.

AI 总结：本文针对不平衡数据集与开放集场景，从理论上重新构建神经崩溃，证明类别平均损失达最小值时标准神经崩溃仍存在，并分析重新平衡的对比损失下的神经崩溃特性。

链接：https://arxiv.org/abs/2611.07753

机构：Shenzhen University（深圳大学）； Institute of Image Understanding Research, North Minzu University（北方民族大学图像理解研究所）

作者：Qiufu Li, Huibing Xiao, Linlin Shen, Yongdong Huang

英文摘要：Neural collapse (NC) describes the elegant geometric structure that sample features and classifier vectors exhibit when a classification model's loss reaches its minimum. Class imbalance, however, generally distorts standard NC. In this paper, we theoretically reformulate NC on imbalanced datasets and in open-set settings. Our analysis is based on an optimization problem over class-wise averaged losses, rather than the global averaged losses used in prior NC studies. We prove that when class-wise averaged losses attain their minima, standard NC is achieved on balanced or imbalanced datasets, for both cross-entropy (CE) and binary cross-entropy (BCE) losses. We further extend the analysis to an open-set-inspired scenario, where some classes contain no training samples yet are still assigned classifier vectors in the losses. We find that in this scenario, when class-wise averaged losses reach their minima, standard NC still emerges on the known classes with training samples, while the classifier vectors of unknown classes without training samples overlap with each other and become orthogonal to those of the known classes. Finally, we analyze Softmax- and Sigmoid-based contrastive losses re-balanced by inverse sample frequency, proving that at their minima on balanced or imbalanced datasets, sample features of different classes likewise form standard NC.

AI 总结：EnergyBench 是面向欧洲电力市场的实时竞价门对齐基准，通过冻结信息集运行模型，发现协变量可显著降低基础模型误差，中位数集成表现最优，电网运营商风电光伏预测仍领先。

链接：https://arxiv.org/abs/2611.07784

机构：Vrije Universiteit Brussel（布鲁塞尔自由大学）； Flanders Alliance for Climate Technology (FACT)（佛兰德气候技术联盟）； DTU Wind and Energy Systems, Technical University of Denmark（丹麦技术大学风能与能源系统系）； University of Granada（格拉纳达大学）； Monash University（莫纳什大学）

作者：Evgenii Genov, Ander Martinez Alonso, Christoph Bergmeir

英文摘要： Energy forecasting models are usually compared on backtests and fixed public datasets. Backtests can use data that did not yet exist when the forecast would have been made, and a fixed test set, reused across studies and possibly contained in the pretraining data of foundation models, invites overfitting to the benchmark. EnergyBench evaluates every entry live instead. Each day at 11:30 CET/CEST, thirty minutes before the day-ahead (SDAC) gate closes, every model forecasts the next day's price, wind and solar generation and load on the 15-minute grid for six bidding zones (BE, NL, FR, DE-LU, DK1, DK2). Unlike existing live benchmarks, EnergyBench runs the models itself, on one information set that it freezes at the cut-off and publishes, including three numerical weather prediction models, so that foundation models can be compared with and without the same weather covariates under the actual market deadline. Errors are pooled per series against a seasonal-naive baseline, and models are ranked by a Bradley-Terry rating on the Elo scale with a day-bootstrap interval. The field spans baselines, linear and gradient-boosted models on the frozen weather, eighteen time-series foundation models from eleven families, ensembles, and the grid operators' own forecasts. Over the first thirty-two live days, the covariates reduced the error of each of the ten foundation models run with and without them by 37-51% on matched races (64-73% on wind, 48-61% on solar). Used zero-shot, these models match or outperform a LightGBM model trained on the same covariates. Model size, from 3M to 331M parameters, has a much smaller effect, and median ensembles of the covariate models rank highest. The grid operators' pre-gate forecasts remain ahead on wind and solar, where the best single foundation models have 16-22% more error; on load the gap is 4-6%, within the intervals.

AI 总结：该研究审计了多模态CXR与EHR分类的类别不平衡策略基准测试，发现原记录划分基准存在标签泄露问题，经患者分离评估后，原结论不成立，还提供了泄露感知协议等工具。

链接：https://arxiv.org/abs/2611.07950

机构：Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology (HCMUT)（胡志明市理工大学计算机科学与工程学院）； Vietnam National University Ho Chi Minh City（越南国家大学胡志明市分校）； Applied AI & Data Science Division, AK Technologies Company Limited（AK科技有限公司应用人工智能与数据科学部）

作者：Dang-Khoa Bui, Quang-Huy Tran, Minh-Khoi Nguyen-Bui, Duc-Tuan Ngo, Thanh-Trong Tran, Tuan-Khoi Nguyen, Hoang-Anh Ngo

英文摘要：Benchmarks of class-imbalance strategies for multimodal chest X-ray (CXR) and electronic health record (EHR) prediction are often evaluated on record-level splits of cohorts in which one patient contributes several radiographs. We audit such a benchmark on 29,083 MIMIC-CXR radiographs of 10,479 patients with eight diagnosis targets. First, the record-split benchmark was misleading: a lookup table that copies the label of a development-set radiograph with the same EHR vector nearly matched the best model, and on unseen patients every model fell to the level of predicting every case positive. Second, tracing the labels back to MIMIC-IV explained why: each patient's labels are keyword matches on the diagnosis codes of a single admission, copied, with one EHR vector per patient, to all of the patient's radiographs, so about a third of the radiographs carry labels that differ from the codes of their own admission. Third, under patient-separated evaluation with admission-aligned and rule-refined labels, the initial conclusions do not hold: in the evaluated configurations no imbalance strategy raised the observed mean macro F2 over no augmentation by more than 0.01, tuned tree ensembles showed no consistent advantage over tuned logistic regression, and the released EHR variables stayed at the level of the all-positive baseline, whereas the diagnostic history available at imaging time and image embeddings exceeded it, including, with rule-refined labels, a CheXpert-trained encoder that never saw these radiographs; RAD-DINO had seen 96.6% of them during self-supervised pretraining. Adding either encoder to the diagnostic history gave no statistically significant F2 gain; exploratory AUROC gains were significant with RAD-DINO but not with DenseNet. We provide reusable diagnostics, a leakage-aware protocol and a checklist.

AI 总结：本研究推出XStack基准，评估智能体在PyTorch与JAX生态系统间迁移ML代码仓库的能力，测试显示现有智能体迁移可靠性不足，该基准为跨框架硬件迁移提供可衡量的研究方向。

链接：https://arxiv.org/abs/2611.08472

机构：Princeton University（普林斯顿大学）

作者：Linrong Cai, Taiming Lu, Haozhe Chen, Zhuang Liu

英文摘要：ML code should run where researchers choose, not where it was first written. Today, machine learning code mainly lives in two ecosystems, PyTorch on NVIDIA GPUs and JAX on Google TPUs. Migrating repositories between them means building and testing across frameworks and hardware. This migration is (1) a practical need, since it unlocks different hardware for researchers, and (2) a natural benchmark: it involves different frameworks and hardware, covers whole repositories, and is long-horizon and challenging. We introduce XStack, a benchmark that evaluates agents on migrating ML repositories between these two ecosystems. Each agent receives a PyTorch repository and a clean TPU machine and must deliver a JAX repository. Its output passes if it meets our grading criteria, based on the source repository's published result, authenticated and calibrated by us. Migration by agents remains unreliable: Claude Opus 5 passes 75%, GPT-5.6-sol 45%, and every other agent at most 40%. XStack is a step toward freeing ML code from its ecosystem. We release XStack to make progress toward reliable migration across frameworks and hardware measurable.

AI 总结：该研究提出分数阶物理信息神经网络，结合符号回归建模社交媒体病毒传播，通过完全白化残差解决噪声问题，将24小时窗口分段后，在348条X帖子上的轨迹重建性能优于经典基线。

链接：https://arxiv.org/abs/2611.07215

机构：Brown University（布朗大学）

作者：Alex Koong, George Em Karniadakis

英文摘要：Existing models of social media virality either prescribe the functional form of engagement growth or assume a fixed mechanism by which past activity influences future engagement. We present a fractional physics-informed neural network that learns both the temporal memory and nonlinear growth dynamics of social media engagement, using symbolic regression to translate the learned dynamics into interpretable equations. Synthetic experiments at the deployment resolution of 24 hourly observations demonstrate that the fractional order and reaction law can be jointly recovered from clean data. Under observation noise, however, the standard pointwise physics residual drives the fractional order toward its lower bound through a scale artifact. We address this failure using a fully whitened residual that propagates observation noise through the fractional operator. Applying the framework to 348 posts from X (formerly Twitter), we find that a single-window model estimates strongly memory-dependent dynamics but fails to accurately reconstruct held-out trajectories. Partitioning the 24-hour window into four teacher-forced segments substantially improves performance, yielding segment-specific fractional orders that increase from 0.23 to 0.99 and interpretable growth laws for each phase. The segmented model reconstructs 64% of held-out trajectories within a mean squared error of 0.01, compared with 44% for the strongest of nine classical baselines. Together, these results demonstrate the potential of fractional physics-informed equation discovery to uncover interpretable engagement dynamics while highlighting the importance of noise-aware estimation and temporal segmentation.

AI 总结：该研究针对初级保健EHR模型，首次系统评估了长达30年的时间上下文对结直肠癌预测的影响，发现18年时判别能力达峰值，还揭示了RoPE在长时序下的混叠问题并提出调整方案。

链接：https://arxiv.org/abs/2611.07265

机构：Imperial College London（帝国理工学院）； Sorbonne Université（索邦大学）； AP-HP（巴黎公立医院集团）

作者：Achille Raviez, Lan Wang, Brendan Delaney

英文摘要：EHR foundation models read medical events within a fixed context window, and long context has so far been measured only in tokens. How far back in time a model must look to capture the earliest predictive signal is unknown. We introduce time-context, the calendar lookback window before the prediction point, and provide, to our knowledge, the first systematic evaluation of decade-scale time-context for an EHR foundation model. We pretrained a decoder-only transformer (34.6M parameters) on 1.4B coded events from 3.6M primary-care patients in North West London and fine-tuned it to predict incident colorectal cancer under a 12-month pre-diagnostic buffer. A frozen cohort of 1,008 cases and 9,582 matched controls, all registered for at least 30 years, was evaluated at 15 nested lookback windows from 2 to 30 years, so that every truncation is compared within the same patients. Discrimination rose from an AUROC of 0.551 at 2 years to 0.590 at 18 years, then plateaued, with about 80% of the gain reached within 10 years; an XGBoost baseline converged to the same level (peak AUROC 0.589). A simulation further suggests that rotary position embeddings (RoPE) over elapsed time alias beyond about 20 years at the default base frequency, motivating a retuned $\theta$. Time-context is a measurable design axis for clinical prediction models, distinct from token-context, and should be reported.

AI 总结：本研究针对工业4.0中钻头卡滞的早期识别问题，提出基于超声波传感器的声学异常检测系统，通过半监督Autoencoder模型在公开数据集上实现了优异的故障检测性能，为工业安全提供了可靠方案。

链接：https://arxiv.org/abs/2611.07356

机构：Federal Rural University of Pernambuco（伯南布哥联邦农村大学）； Western Sydney University（西悉尼大学）

作者：Maria Beatriz Silva, Lucas Edson Silva de Araújo, Sergio F. Chevtchenko, Ermeson Andrade, Danilo Araujo, Saeed Afshar

英文摘要：In the context of Industry 4.0, predictive maintenance is essential to minimize unplanned downtime and ensure operational safety. In machining processes, drill jamming is a critical failure that compromises both workpiece quality and tool integrity. This work proposes an acoustic anomaly detection system using ultrasonic sensors to identify pre-failure patterns in Computer Numerical Control (CNC) machining. We release a public dataset of time- and frequency-domain acoustic features from 17 drills, and benchmark semi-supervised autoencoder-based normality modelling against supervised XGBoost classification using Leave-One-Drill-Out validation. The results demonstrate that the Autoencoder is highly effective for industrial safety, achieving a Recall of 1.00 and a consolidated F1-score of 0.81. Furthermore, the threshold-independent analysis shows that the Autoencoder better separates normal and pre-failure acoustic patterns, achieving a consolidated AUC of 0.8294, compared with 0.7559 for the supervised model under inter-tool covariate shift. In contrast, the supervised XGBoost achieved an average F1-score of 0.55 and failed to detect failures in some of the evaluated drills. The findings highlight that the ultrasonic spectrum provides a reliable window for identifying incipient mechanical instabilities, and the AE model offers a robust safety envelope by anticipating all jamming events across the tested drill bits.

AI 总结：该研究针对时间序列预测中自生成误差会扰动未来预测的问题，提出残差导向局部展开一致性正则化方法，通过惩罚雅可比向量积抑制误差回声，提升了预测精度。

链接：https://arxiv.org/abs/2611.07473

机构：Bosch Center for Artificial Intelligence (BCAI)（博世人工智能中心）； Bosch (China) Investment Co., Ltd.（博世（中国）投资有限公司）

作者：Zheng Wang, Kaixuan Zhang, Xiaonan Lu, Wanfang Chen

英文摘要：In time-series forecasting, rollout errors are not merely one-step fitting discrepancies: once fed back into the evolving state, they can perturb future predictions. We formalize this effect through local rollout consistency, which compares the next prediction produced under teacher forcing with that produced under free running. In principle, this consistency can be enforced directly by tracking the residual-induced perturbation of the next state. In practice, however, the residual-to-state alignment operator is often difficult to design when the input state and output block have different dimensions, heterogeneous components, or architecture-dependent layouts. We therefore use a first-order local surrogate that penalizes the Jacobian-vector product of the predictor along the actual feedback perturbation, suppressing precisely the directions through which self-generated errors remain visible to future predictions. We further give the same loss a geometric interpretation near a predictive manifold: by controlling residual-induced directions that probe the normal bundle, the objective yields theoretical conditions under which free-running trajectories remain inside a stability tube around the manifold. Experiments on time series forecasting tasks support this view, showing that the proposed regularization improves predictive accuracy. More broadly, our results suggest a training principle for time-series forecasters and sequential predictors: a model should not only make small one-step errors, but also learn not to let its own errors echo into the future. Code is available at this https URL.

AI 总结：本研究针对零样本多变量时间序列异常检测，提出将大语言模型仅作为特征提取器，通过孤立森林评估特征可分性，为跨模态基础模型的异常检测应用提供实证支持。

链接：https://arxiv.org/abs/2611.07720

机构：University of Luxembourg（卢森堡大学）

作者： Romain Hermary, Nesryne Mejri, Djamila Aouada

英文摘要：Time-Series Anomaly Detection (TSAD) is a fundamentally challenging task: it is unsupervised, requires high local precision capabilities, and suffers from limited labelled data. Recent work has explored adapting Large Language Models (LLMs) to time-series via cross-modal transfer. However, existing approaches are poorly suited to Multivariate TSAD (MTSAD), as tokenisation often obscures fine-grained temporal and variables structures, and most methods optimise auxiliary objectives such as forecasting or reconstruction. We posit that these surrogate objectives become unreliable as model capacity increases. Highly expressive models risk reconstructing or forecasting anomalies, undermining the foundations of both strategies. Instead, we propose to use LLMs purely as feature extractors and directly analyse their induced feature spaces. In this work, investigate the ability of LLMs to provide useful representations for MTSADs in a zero-shot setting through classification. To this end, we study the separability of the extracted features, under the assumption that anomalous features should be isolated from normal ones. We analyse their temporal and contextual discriminative ability using simple step-wise tokenisation schemes, with one time-step per token, which preserve fine-grained local information. Their usability for MTSAD is assessed using an unsupervised Isolation Forest, enabling a direct evaluation of the resulting representation quality. We further investigate whether the observed performance can be attributed to the language-related pre-training of the LLMs or arises independently of it. Our results provide empirical insight into the ability of LLMs to represent multivariate time-series for anomaly detection and highlight broader applications of cross-modal foundation models.

AI 总结：该研究提出FRPO方法，通过联合监督资产机会与交易结果强化投资智能体决策，在五个投资组合设置中实现最高平均累计与年化收益。

链接：https://arxiv.org/abs/2611.07750

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

作者：Zeyu Liu, Xuanzhi Feng, Haoran Xu, Lai Sing Kwong, Yang Li, Hualei Zhang, Hao Chen, Jie Zhang, Xiaoyi Pang, Song Guo

英文摘要：Online investment services must translate asset opportunities into investment outcomes for users' accounts, yet the same market opportunity can warrant different account actions. We study how supervision from future asset opportunities can strengthen account-level trading decisions and propose Forecast-Augmented Relative Policy Optimization (FRPO). FRPO separately supervises asset opportunities and account decisions through a jointly learned representation: a Financial State Head (FSH) learns future relative-return targets to provide asset-level opportunity supervision, while Same-Path Group-Relative Optimization compares trading trajectories under the same market path and initial account to provide account-level relative-outcome supervision. An Adaptive Gate coordinates auxiliary supervision using predictive variance and gradient alignment. FSH is used only during training; inference uses the encoder and policy for trading. Across DJIA30, TOPIX Core30, and three investment-style portfolios, ten-seed evaluations show that FRPO achieves the highest mean cumulative and annualized returns among the compared methods in all five settings. These results support strengthening investment agents' trading decisions through joint supervision from asset opportunities and trading outcomes.

AI 总结：本研究通过模拟实验发现，大型语言模型审计临床预测网络的关键是计算与提示引导，而非权重访问，黑盒访问效果与白盒相当，明确指令可减少其弃权情况。

链接：https://arxiv.org/abs/2611.07847

机构：Tianjin Medical University（天津医科大学）； Tianjin Medical University General Hospital（天津医科大学总医院）

作者：Zhiyu Liu, Xinyu Yang

英文摘要：Frontier large language models (LLMs) have been reported to read, edit and write the weights of small neural networks, raising the prospect of auditing clinical prediction models from their files alone. Public demonstrations, however, mixed reading raw weights with running code. We separated these conditions in internally pre-specified simulation experiments. Tanh networks of 105 to 38,401 weights were trained on simulated cohorts of patients with subarachnoid haemorrhage, each with a planted relationship or none, and one network was trained on labels that under-recorded deaths in a subgroup. GPT-6 Astra and four other GPT models answered single prompts under a default or an explicit system instruction, either with the weights in the prompt and no tools, or with a calculation-only tool that exposed the weights (white box) or only the network's predictions (black box). Without tools, Astra recovered planted relationships in 353-weight networks but became unreliable at 3,457 weights, and under the default instruction it declined 11 of 12 audit and repair tasks as unidentifiable; the explicit instruction removed these refusals. With the tool, it recovered the planted relationship in all 29 runs on networks of up to 38,401 weights, raised no false alarm in 15 runs on null networks, and found the biased subgroup in 6 of 6 runs. Black-box access performed as well as white-box access, and other GPT-6 models read as accurately. A well-designed script matched the comparator models on the audit and the best LLM repair. For auditing these networks, what mattered was computation and elicitation, not access to the weights.

AI 总结：PersonaAgent是一种智能体式交易异常检测框架，通过可刷新多实体角色记忆解决交易模式高频变化与实体嵌入表问题，在Visa基准测试中高效提升了交易异常评分的PR-AUC及生产业务指标。

链接：https://arxiv.org/abs/2611.08094

机构：Case Western Reserve University（凯斯西储大学）； Visa Research（维萨研究中心）

作者：Zhengyu Fang, Liang Wang, Yingtong Dou, Yuzhong Chen, Shubham Jain, Junpeng Wang, Xiran Fan, Menghai Pan, Jiarui Sun, Yan Zheng, Yiwei Cai

英文摘要： In large-scale payment networks, entities (card, merchant, etc.) and their interactions play a key role in transaction anomaly detection. However, transaction patterns between different entities usually change more frequently than the production model refresh schedule. Meanwhile, entity embedding tables suffer from high-cardinality and sparse interaction issues. To resolve the above challenges, we present PersonaAgent, an agentic transaction anomaly detection framework which represents the latest entity behavior signals as personas and saves them in a refreshable memory. Offline, a weekly scheduled batch job is used to extract the latest behavior signals and update them in the memory. At inference time, a lightweight adapter retrieves the corresponding entity behavior signals of each transaction from memory and uses them to augment the base anomaly classifier's scoring. Through comprehensive benchmarking with different anomaly detection models at Visa, PersonaAgent is able to score transactions efficiently (a 1,000-transaction CPU batch in 1-2 ms), lifting the PR-AUC by 0.65-6.58% and the production business metric by 8.93-14.24% relatively in deployment.

AI 总结：该研究提出PRISM框架，结合固定随机非线性特征与马氏距离评分，在未优化权重或使用预训练表征的情况下，于表格和视觉异常检测任务中取得优异性能，可作为评估学习表征价值的基线。

链接：https://arxiv.org/abs/2611.08136

机构：Sejong University（世宗大学）； Korea Atomic Energy Research Institute（韩国原子力研究院）

作者：Hyojun Yi, Geunyoung Kim, Hogeon Seo, Seunghyoung Ryu

英文摘要：Anomaly detection (AD) is often framed as a representation-learning problem, where strong performance is expected to require learned features or AD-specific optimization. However, this view can overlook the intrinsic separability of anomalies from normal references, even in the absence of learned representations. We therefore investigate how far untrained neural networks can go in anomaly detection. To this end, we propose PRISM, a framework combining fixed random nonlinear features with normal-reference Mahalanobis scoring for tabular detection and dense visual localization. Across 47 tabular datasets in ADBench, PRISM-T achieves the highest mean AUROC (86.4%) and the second-highest mean AUPRC (65.9%) among the compared methods. On MVTec AD and VisA, covering 27 visual categories, PRISM-V achieves mean image-level and pixel-level AUROCs of 90.4% and 96.4%, respectively. Notably, both models achieve these results without neural weight optimization or pretrained representations. A theoretical analysis further characterizes how nonlinear transformations affect the mean and covariance contributions to the Mahalanobis score gap, while empirical analyses examine the resulting score separability. These findings establish PRISM as both a competitive anomaly detector and a prior-to-learning baseline for assessing the added value of learned representations in anomaly detection.

AI 总结：该研究针对时变偏微分方程场景下共形预测覆盖保证失效的问题，通过理论分析指出函数空间中分布相互奇异、离散场景中总变差距离指数饱和的特性，并提出加权共形预测方法以获取有限样本覆盖保证。

链接：https://arxiv.org/abs/2611.08409

机构：LMU Munich（慕尼黑大学）； MCML（慕尼黑计算与机器学习中心）； TU Munich（慕尼黑工业大学）； Helmholtz AI（亥姆霍兹人工智能研究所）； relAI（可靠性人工智能中心）； DFKI（德国人工智能研究中心）

作者：Joshua Stiller, Annika Schneider, Eyke Hüllermeier

英文摘要：Uncertainty quantification is crucial in scientific machine learning, where models inform safety-critical tasks such as flood forecasting, financial risk management, and thermal control in machines. Conformal prediction provides distribution-free coverage guarantees, but in time-dependent settings common to physics and engineering, these guarantees can break down, leading to systematic undercoverage. We study this problem in the context of surrogate models for time-dependent physical systems described by partial differential equations. We prove that in a function space setting, distributions at arbitrarily close times can be mutually singular, making exact coverage guarantees impossible. We further show that in discretized settings, the total variation distance between distributions often saturates exponentially fast with the grid resolution, mandating restrictions on the spatial discretization to ensure coverage. Finally, for certain settings, we show how to use principled weighted conformal prediction to obtain finite-sample coverage guarantees over growing time horizons.

AI 总结：该研究提出一种数据驱动方法，从微分方程数据中发现内在李点对称性，结合延拓、子空间限制等技术，在11维对称代数基准上成功恢复生成元及结构，可从微分系统数据中恢复显式连续对称生成元。

链接：https://arxiv.org/abs/2611.08421

机构：Middle East Technical University（中东技术大学）； ASELSAN Inc.（阿塞勒桑公司）； ROMER Robotics and AI Center, Middle East Technical University（中东技术大学ROMER机器人与人工智能中心）

作者：Ilker Gurcan, Kadir Ustun, Ramazan Gokberk Cinbis

英文摘要：Discovering continuous symmetries directly from differential-equation data can reveal geometric structure for model reduction and scientific machine learning without symbolic determining equations or a closed-form representation of the governing system. We formulate intrinsic Lie point-symmetry discovery as an operator-theoretic problem on sampled jet-space data. Admissible generators are identified from a sampled determining operator and separated from redundant, solution-dependent, and geometrically trivial directions. The framework combines prolongation, sequential subspace restrictions, and a fixed nonlinear dictionary linear in trainable coefficients. A rank-stabilization preflight test diagnoses systems whose represented intrinsic symmetry algebra does not stabilize under dictionary refinement. On a controlled differential-equation benchmark with an analytically known 11-dimensional represented intrinsic symmetry algebra, the method recovers the complete represented generator set, reproduces its Lie-bracket structure to numerical precision, and approximates a known generator beyond the training domain. Here we show that explicit intrinsic continuous-symmetry generators can be recovered from data obtained from exact or approximate differential systems.

AI 总结：本研究提出SteelAgent模型，结合冶金描述符、机器学习与大语言模型实现钢性能预测，通过实验揭示钢性能模型在不同划分方式下泛化与不确定性的差异，为评估报告提供依据。

链接：https://arxiv.org/abs/2611.06953

作者：Aleksandr Volkov, Roman Sultimov, Mikhail Kuzin, Yury Maximov

英文摘要： Steel property models can perform well on random test splits even when they extrapolate poorly to unfamiliar alloys. SteelAgent couples metallurgical descriptors and machine-learning models with a language-model interface for property prediction, uncertainty reporting and inverse design. We examine how the choice of held-out unit changes the conclusions drawn from SteelBench tensile-strength evaluations. For gradient boosting, pooled out-of-fold $R^2$ fell from 0.859 on a random split to 0.113 when steel families were held out. Empirical coverage of nominal 90% prediction intervals ranged from 91.0% under random splitting to 37.6% on held-out sources. The effect of physics-derived features depended on the evaluation. They increased gradient-boosting $R^2$ by 0.2534 in a 17-family holdout analysis and decreased it by 0.358 in a separate four-grade laboratory analysis. We report group-level errors and a synthetic regression experiment illustrating how calibration can deteriorate on held-out groups. These results motivate reporting the evaluation population alongside a property estimate or compliance score.

AI 总结：该研究在MONSTER基准的六个数据集上测试Reverso、Chronos-Bolt等冻结预测模型的分类性能，发现其预训练表示未可靠优于非预训练基线，仅在部分数据集有小幅收益，强调需重视任务特定基线与随机权重对照。

链接：https://arxiv.org/abs/2611.06988

机构：Dwarkadas J. Sanghvi College of Engineering, University of Mumbai（孟买大学德瓦卡达斯·J·桑格维工程学院）

作者：Rishit Jain, Rishit Kar, Chetashri Bhadane, Nilesh Patil

英文摘要：Forecasting foundation models are increasingly used as frozen feature extractors for time series classification, but it remains unclear whether their pretrained representations offer an advantage over strong classifiers that do not require pretraining. This raises an important question: do these representations provide useful information beyond what the model architecture alone can capture, particularly when labelled data are scarce? We present a controlled benchmarking of two forecasting models, Reverso and Chronos-Bolt, on six large datasets from the MONSTER benchmark, using 1%, 10%, and 100% of the available training labels. We compare their frozen representations against four established time series classification methods and random-weight versions of the same forecasting models. Our results show that pretraining improves classification accuracy over random-weight counterparts on five of the six datasets, with gains of up to 23 percentage points. However, the frozen representations do not reliably outperform the strongest non-pretrained baseline in any setting: they perform worse in 15 of 17 settings and tie in the remaining two. Further experiments show that restoring amplitude information recovers part of the performance gap, while fine-tuning and forecast-error features do not close it. These findings highlight the importance of strong task specific baselines and random weight controls when evaluating forecasting foundation models for classification. Our code is available at this https URL.

AI 总结：该研究针对数据流的概念漂移问题，提出时间中性树（TNT）决策树架构，通过中性化旧样本并按需重新纳入，在USP DS仓库的24个数据流上验证其获同类最佳性能，实现鲁棒流分类。

链接：https://arxiv.org/abs/2611.06991

机构：Utrecht University（乌得勒支大学）

作者：Brandon Gower-Winter, Georg Krempl

英文摘要：Data Streams are characterised as a potentially infinite source of data subject to non-stationarity. This non-stationarity, also called Concept Drift, often results in deployed predictive models having to be perpetually retrained as the underlying data generating process of the stream changes over time. Drift Detectors are a class of methods used to identify these changes, triggering these retraining procedures. Historically, Stream Learning has operated under the assumption that instances are only seen once, and cannot be stored for later use. Contemporary work relaxes this assumption, yet drift detection and simple resetting procedures are still ubiquitous. Building on related work which has shown that traditional Batch Learning algorithms often outperform Stream Learning algorithms when instances can be stored, we introduce Time Neutralising Trees (TNT), a Decision Tree architecture that enables robust stream classification in settings with Concept Drift. During training, TNT will filter out ("neutralise") old training samples that are no longer relevant given the current context of the stream. If those instances become relevant again, TNT will then re-include them. We evaluate TNT, and related algorithms on 24 real and semi-real classification data streams from the USP DS repository. Results show strong evidence that TNT achieves best-in-class performance on a wide-range of data streams with varying concept drift types.

AI 总结：本文从随机动力学系统视角，将LLMs构建为高维非线性自回归模型，刻画交互LLM网络的集体行为，提出神经SDEs作为连续时间生成模型，为生成式AI研究提供了新的理论框架。

链接：https://arxiv.org/abs/2611.07065

机构：Cornell University（康奈尔大学）； Yale University（耶鲁大学）

作者：Vikram Krishnamurthy, Yiming Zhang, Luke Snow

英文摘要：This paper develops a stochastic dynamical-systems perspective on three classes of generative AI models. First, we formulate large language models (LLMs) as high-dimensional nonlinear autoregressive models with attention-based dependencies. This formulation connects pretraining by conditional maximum likelihood to sequential token generation and provides a probabilistic interpretation of LLM operation. Second, we consider networks of interacting LLMs and characterize their collective behavior through mean-field ordinary differential equations. These dynamics provide a framework for understanding emergent preferential attachment and persistent structural biases, including type-dependent glass-ceiling effects. Third, we consider neural stochastic differential equations (SDEs) as continuous-time generative models for stochastic trajectories, including those observed at irregular time intervals. We discuss their training with particular emphasis on Wiener-space cubature methods based on rough-path analysis as an efficient alternative to Monte Carlo estimation. Together, the three settings address probabilistic prediction, information exchange and emergent behavior in interacting networks, and path-space generative modeling.

AI 总结：该研究提出Mechanistic Shaping框架，在训练中通过模型内部表征塑造模型以遵循概念规范，发现其可减少目标行为且保留任务性能，还明确了概念规避监控的机制及监控可行性，为基于可解释性的训练提供了方向。

链接：https://arxiv.org/abs/2611.07071

作者：Thomas Fel, Matthew Kowal, Matt Feiszli, Satchel Grant, Andrew Lee, Chris Earls, Fatih Dinc, Vasudev Shyam, Usha Bhalla, Leon Bergen, Siddharth Boppana, Ching Fang, Michael Byun, Tristan Gallent, Nicholas Wang, Tal Haklay, Dron Hazra, Neehar Kondapaneni, Connor Watts, Ryan Panwar, Thomas Dooms, Jack Merullo, Atticus Geiger, Ekdeep Singh Lubana, Thomas McGrath, Owen Lewis

英文摘要：We propose Mechanistic Shaping, a unified framework for shaping models through their internal representations. Mechanistic Shaping is applied during training so that models follow specifications (rules) about which concepts they express, measured via their outputs or internal representations (e.g., no harmful content), and also which concepts they rely on, i.e., what they use to make their decisions (e.g., not a candidate's gender when screening a résumé). Across multiple domains (vision, medical imaging, language), scales (up to 32B parameters), and training paradigms (classification, SFT, DPO, GRPO), we find that Mechanistic Shaping substantially reduces targeted behaviors while preserving task performance. Since training against a representation can also modify it, we ask whether shaped concepts remain monitorable, a question that depends on the class of monitors considered. We therefore introduce a hierarchy of such classes, which separates a concept that escapes one probe from a concept that is truly gone. We characterize three geometric mechanisms of circumvention: the concept can drift away from the monitor, be erased from the layer, or persist in higher-order statistics as a nonlinear encoding, which we explain through a product-code construction. Although other mechanisms may arise in different settings or at larger scales, these three account for all of our observations, and each can be detected by a matching monitor. In practice, we find that shaped concepts largely remain monitorable: refitted linear probes recover the concept in most settings, and the original probe remains reliable whenever gradients do not flow through it. Taken together, these results point to the potential of using interpretability as a tool for training, and call for a correspondingly rigorous science of interpretability-based training.

AI 总结：该研究探讨主动信息获取中深度与描述的关系，提出用编码决策的合适潜在原因的一步分数替代高成本的精确前瞻策略，可在极小搜索量下替代大部分深度，同时指出语言模型概率迁移的相关问题。

链接：https://arxiv.org/abs/2611.07155

机构：JPMorgan Chase & Co.（摩根大通公司）

作者：Biswa Sengupta

英文摘要：A proactive agent must decide what to ask before it acts. Exact lookahead values combinations of questions but is expensive and leans on a fitted model; a one-step score is cheap but can miss them. Our main finding: the need for depth often lies in how the hidden cause is described. A one-step score on the label can be blind to an interaction among answers. A score on a suitable latent cause is not, and where that latent encodes the decision it replaces most of depth at a fraction of the search. Described differently, the same problem can defeat the score while exact values do not move, so keep the latent small and check it against exact values. Probabilities learned from one language model's answers do not carry to another unless planning allows for both; those read from tokens are worth only a few labels.

AI 总结：本综述梳理了数据同化领域的非线性滤波方法，重点介绍了集成得分滤波器（EnSF）的基础与进展，分析其优势及14种扩展方法的特点，确立其为非高斯高维滤波的最先进方案。

链接：https://arxiv.org/abs/2611.07160

机构：Florida State University（佛罗里达州立大学）

作者：Cath Pham, Kadesha Reynolds

英文摘要：Many real-world applications, such as numerical weather prediction or wildfire tracking, involve modeling highly nonlinear dynamical systems that are chaotic and influenced by various sources of uncertainty. To improve model prediction accuracy and reduce such uncertainties, data assimilation (DA) combines the dynamical model with partial, noisy observations to estimate the evolution of the system's state. Doing so depends on solving the optimal filtering problem: computing the conditional distribution of a hidden state given a sequence of observations. Classical approaches fail along three distinct axes: Kalman-based methods assume Gaussian posteriors and produce biased estimates under nonlinear observation operators; particle-based methods represent distributions with weighted samples and collapse in high dimensions due to the curse of dimensionality; and kernel-based methods move beyond discrete particle representations but cannot maintain their density approximation affordably as dimension grows. The Ensemble Score Filter (EnSF) avoids all three constraints by encoding the filtering density in a score function and approximating that score with a training-free Monte Carlo estimator, enabling stable filtering in high-dimensional, chaotic systems. This survey traces EnSF's mathematical foundations, establishes it as the state of the art for non-Gaussian, high-dimensional filtering, and examines fourteen extensions that expand its operational reach. Twelve of the fourteen extensions preserve this training-free Monte Carlo estimator at the core of the update, with only two departing from it to achieve planetary-scale deployment.

AI 总结：该研究提出SMToM框架，在BDI结构中从轨迹归因显式心理状态，实例化的BDIBottleneck模型在行人导航任务中优于基线，支持可解释的反事实分析。

链接：https://arxiv.org/abs/2611.07208

机构：Halıcıoğlu School of Data Science and Computing, University of California San Diego（加州大学圣地亚哥分校哈利西奥格鲁数据科学与计算学院）； New York University（纽约大学）； Department of Philosophy and School of Data Science, University of Virginia（弗吉尼亚大学哲学系与数据科学学院）

作者：Nigel F. Doering, Rahath Malladi, Arshia Sangwan, M Tanjid Hasan Tonmoy, Jacob Doan, David Danks, Tauhidur Rahman

英文摘要： Predictive models of human behavior trained on large-scale trajectory data optimize for statistical accuracy without representing the mental states that causally generate behavior. Such models support prediction but not principled intervention: they cannot answer how an agent's behavior would change if its beliefs or preferences were different. We introduce Structured Machine Theory of Mind (SMToM), a framework that attributes explicit, independently supervised belief and desire representations from observed trajectories within a Belief-Desire-Intention (BDI) causal structure. The central architectural element is a goal head that consumes only the predicted mental-state channels and a current-trajectory embedding; counterfactual intervention on beliefs and desires is a direct operation. We instantiate SMToM on a controlled pedestrian navigation domain where ground-truth mental states are known by construction, enabling evaluation of attribution accuracy, local channel interventions, and trajectory-level counterfactual diagnostics. The resulting model, BDIBottleneck, improves over trajectory-only and unsupervised-context neural baselines. It is also competitive with a matched auxiliary-supervised context baseline while providing an editable mental-state interface that the unbottlenecked baseline lacks. Counterfactual experiments on desires confirm that substituting an agent's inferred preferences with a different activity type coherently shifts predicted destinations toward relevant locations. Counterfactual experiments on beliefs confirm that marking a location as unavailable in the agent's belief state reduces its predicted probability as a destination, though in a complex way. These results demonstrate, in a controlled navigation setting, that explicit BDI-structured supervision is a viable foundation for interpretable and counterfactual analysis of longitudinal trajectory data.

AI 总结：TileQuant是一种无需训练的LLM混合精度后训练量化框架，将精度分配给硬件对齐的二维权重Tile，可提升吞吐量并在多任务中保留性能且优于相关模型。

链接：https://arxiv.org/abs/2611.07217

机构：University of North Texas（北得克萨斯大学）； Saint Louis University（圣路易斯大学）

作者：Hanzhi Zhang, Qiao Zhang, Qinglei Cao, Heng Fan, Yan Huang, Kewei Sha, Yunhe Feng

英文摘要：Post-training quantization (PTQ) reduces large language model (LLM) deployment costs, but preserving quality requires accounting for uneven weight sensitivity. Uniform PTQ uses one weight bit-width, importance-aware PTQ adjusts scales or reconstructs weights within that width, and column-wise mixed precision retains selected columns at higher precision. However, sensitivity varies within columns, so whole-column retention also protects less sensitive weights, while runtime outlier detection adds decoding overhead. We therefore introduce TileQuant, a training-free framework that assigns precision to hardware-aligned 2D weight tiles. Offline calibration provides input-activation-aware scoring from forward statistics for lower calibration cost and gradient-informed scoring from loss sensitivity for stronger quality preservation. Within each projection's retention budget, the highest-scoring tiles remain at 16-bit and the rest use 8-bit. The resulting map guides reconstruction into compact weight buffers, which the kernel reuses for native integer and 16-bit computation with local accumulation to reduce memory traffic. Experiments show that TileQuant reaches 2.66x the original model's generation throughput. Across language modeling, reasoning, long-context retrieval, and question answering, it preserves quality and surpasses quantized and original models on multiple tasks. These results support selecting precision units that capture local sensitivity while enabling efficient hardware execution. The implementation is: this https URL

AI 总结：该研究提出Loopback Transformer，通过残差连接的深度-记忆视角设计，在减少参数的同时，在ClimbMix数据集及13个下游任务上取得优于标准Transformer和其他回环残差设计的性能。

链接：https://arxiv.org/abs/2611.07226

机构：University of Wisconsin–Madison（威斯康星大学麦迪逊分校）； University of California, Santa Barbara（加利福尼亚大学圣巴巴拉分校）； California Institute of Technology（加州理工学院）； Microsoft（微软公司）

作者：Xuan Ouyang, Zefan Cai, Chenlong Zhang, Cheng Luo, Ke Wan, Junjie Hu

英文摘要：Looped Transformers reduce parameter count by repeatedly applying shared blocks, but parameter sharing alone does not determine how the results of repeated computation are reused. We take the depth-memory view of residual connections and analyze which representation is stored after each block application and how later reads weight it. A full block output contains both its input and its new update, so weighting that output also reweights the history already carried by the input. Reading individual updates instead, with the shared stack's input kept on a separate path, gives separate controls over input strength, total update weight, and allocation across updates. The Loopback Transformer implements these controls: geometric decay sets the total update weight and supplies a recency prior, and attention learns how to allocate that weight. On ClimbMix, Loopback reaches lower validation loss than standard Transformer baselines with 40-75% of their Transformer parameters (excluding token embeddings and the output head) at reference scales from 191M to 3B, with training tokens and executed depth matched. It also has higher mean accuracy over thirteen downstream tasks at 1.4B and 3B. At 191M, 374M, and 778M, it has the lowest validation loss and the highest thirteen-task mean among the standard Transformer and seven looped residual designs.

AI 总结：该研究利用神经网络的表示自由度，通过精确函数保持变换与受控功能扰动，结合物理压缩器选择廉价表示，在SmolLM2和Qwen3上使Zstandard压缩检查点大小降低最多23.4%，证明其为实用存储资源。

链接：https://arxiv.org/abs/2611.07233

作者：Toshiaki Koike-Akino

英文摘要：Neural networks admit many parameterizations of the same or nearly the same function, but deployment typically stores whichever parameter vector training happens to produce. We show that this representational freedom can itself be used for compression. Rather than redesigning the codec, we search over exact function-preserving transformations and controlled functional perturbations, using the unchanged physical compressor itself to select cheaper representations. Exact gauge families provide a zero-distortion reference under the strict audit; allowing small functional changes exposes much larger rate-distortion gains. Across SmolLM2 and Qwen3, our method reduces Zstandard-compressed checkpoint size by up to 23.4\%, with similar relative reductions under gzip, Brotli, and ZipNN, establishing trained-network representational freedom as a practical storage resource.

AI 总结：$D^2$-OPD将外推式策略蒸馏的方向与距离解耦，通过硬KL约束限制目标距离、自适应组合确定方向，在三种场景的数学推理基准中比最强基线最高提升9.45分。

链接：https://arxiv.org/abs/2611.07241

机构：Columbia University（哥伦比亚大学）； AWS AI Lab（亚马逊云科技AI实验室）； Amazon AGI（亚马逊通用人工智能实验室）

作者：Bowen Fang (Columbia University), Dingmin Wang (AWS AI Lab), Hengzhi Pei (Amazon AGI)

英文摘要：On-policy distillation (OPD) trains a student model to match a teacher's next-token distributions on student-generated prefixes. Extrapolative OPD further extends the teacher's log-probability shift relative to a reference policy to construct stronger learning targets. However, existing methods couple extrapolation direction and distance, allowing the magnitude of the teacher's shift to dictate how far the target moves. We propose $D^2$-OPD, decoupling these two factors: a per-prefix hard KL constraint bounds the target's distance from the teacher, while an adaptive combination of the teacher's post-training shift and the initial student--teacher gap determines the extrapolation direction. We optimize the student toward this target using forward KL, providing learning signals even for low-probability tokens. We evaluate three settings: where standard OPD performs well, where it struggles with base-model initialization, and where no teacher reference is available. On mathematical reasoning benchmarks, $D^2$-OPD attains the highest score in all three settings, outperforming the strongest baseline by up to 9.45 points. Ablations validate the contributions of each component.

AI 总结：该研究提出KGN方法，通过知识图谱三元组调整嵌入距离重构邻域图，结合IDP分析残差，在8个数据集-编码器条件下提升了嵌入可视化的NMI和LT_2D，缩小了嵌入邻域与语义关系的差距。

链接：https://arxiv.org/abs/2611.07246

机构：DATUMO Inc.（DATUMO公司）

作者：Chang Won Kim, Jaewook Lee, Haeun Min, JunNyeong Jeong, Minwoo Sim, Barom Kang, Kyohoon Jin

英文摘要：In embedding visualizations, nearby instances are interpreted as semantically related, although neighborhoods derived from input embeddings may not reflect content relationships. Neighborhood-based dimensionality reduction (DR) methods can propagate this limitation by optimizing layouts from embedding-based neighborhood graphs. We propose Knowledge-Guided Neighborhood Graph Construction (KGN), which represents relational knowledge derived from each instance as a set of knowledge graph (KG) triples. It uses the degree of triple sharing as pairwise semantic evidence to adjust embedding distances and reconstruct the top-k neighborhood graph; neither KG extraction nor neighborhood construction uses class labels. We also introduce IDProfile (IDP), which interprets residual mismatches by independently assessing whether the source content supports the assigned class label and the neighborhood-majority class. Across eight text and image dataset-encoder conditions, we characterize KGN over the full guidance-strength sweep and four KG similarity measures. Across this sweep, the maximum NMI and LT_2D each exceed the corresponding kNN baseline in all eight conditions, with the largest gains reaching 0.68 and 1.35, respectively. The gains and the extent of neighborhood reconfiguration vary across conditions and are larger under encoder conditions with lower kNN baseline scores. The IDP analysis shows that the largest reductions are concentrated in Placement Issue candidates, whereas Label Issue and Unresolved Case instances account for a larger share of the remaining review set. These results indicate that KGN can narrow the evaluated gap between embedding-based neighborhoods and content-supported semantic relations, while IDP shows that a residual mismatch should not automatically be interpreted as a projection failure.

AI 总结：提出EANK方法，用遗传算法优化实例依赖邻域策略，在多数据集上优于默认k=5，实现了更优的最近邻分类性能。

链接：https://arxiv.org/abs/2611.07257

机构：University of Maroua（马鲁阿大学）； National Advanced School of Engineering, University of Maroua（马鲁阿大学高级工程学院）； Higher Teacher’s Training College, University of Maroua（马鲁阿大学高等师范学院）； Laboratoire de Recherche en Informatique (LaRI)（计算机研究实验室（LaRI））

作者：Isaac Touza, Kaladzavi Guidedi, Kolyang

英文摘要：The number of neighbours k is the central hyperparameter of the k-nearest neighbour (k-NN) classifier and is usually selected globally through cross-validation or evolutionary search. However, the bias-variance trade-off is inherently local: dense and noise-free regions may favour small neighbourhoods, whereas sparse or noisy regions may require larger ones. Since global-k optimization is one-dimensional, replacing exhaustive search with a genetic algorithm offers little computational advantage. We therefore optimize an instance-dependent neighbourhood policy instead. We propose EANK (Evolutionary Adaptive Neighborhood K), which uses a low-dimensional policy to map local neighbourhood descriptors, including density, label entropy, vote margin, and concentration, to an adaptive number of neighbours k(x). A real-coded elitist genetic algorithm optimizes this policy using leave-one-out accuracy with a neighbourhood-smoothness penalty. Precomputed neighbour votes reduce fitness evaluation costs. We investigate theoretical properties of the method, including dominance over global-k solutions on the training criterion, evolutionary convergence, policy regularity, and generalization guarantees. Experiments on synthetic data with spatially heterogeneous structure and twelve public datasets evaluate predictive performance, robustness, and computational cost. Under nested cross-validation, EANK achieves competitive performance relative to tuned global-k baselines and improves upon the default k=5. We also examine the gap between training and test performance to assess the practical value of adaptive neighbourhood selection.

AI 总结：本研究通过系统映射梳理资源受限嵌入式系统中TinyML的能量相关研究，明确其模型、方法、影响因素与优化策略，指出标准化不足问题，为节能TinyML方案开发提供指导。

链接：https://arxiv.org/abs/2611.07264

机构：Federal Rural University of Pernambuco (UFRPE)（伯南布哥联邦农村大学）； Federal University of Alagoas (UFAL)（阿拉戈斯联邦大学）； Western Sydney University（西悉尼大学）

作者： Clariele Almeida, Renan Beserra, Rafael Moura, Danilo Araújo, Bruno Nogueira, Sérgio Chevtchenko, Ermeson Andrade

英文摘要：The rapid expansion of the Internet of Things (IoT) has increased the volume of data generated at the network edge, where many devices operate under severe memory, processing, and energy constraints. In this context, Tiny Machine Learning (TinyML) enables machine learning (ML) tasks to be executed on microcontrollers, smart sensors, and other resource-constrained platforms. However, the literature remains heterogeneous regarding the algorithms evaluated, energy-measurement methods, reported metrics, influencing factors, and optimization strategies. This study presents a systematic mapping of energy-aware TinyML research. It identifies the main models and learning approaches, classifies energy-evaluation methodologies and metrics, examines the technical factors affecting energy consumption, and organizes the reported optimization techniques into a structured taxonomy. The findings highlight the frequent use of convolutional and lightweight architectures, direct measurements and analytical estimations, and factors related to model architecture, hardware, communication, numerical precision, and input characteristics. The optimization strategies were grouped into model compression, hardware-aware design, system and communication optimization, and data and processing optimization. The results also reveal limited standardization in experimental platforms, measurement procedures, and reporting practices, limiting reproducibility and direct comparison across studies. This mapping provides an updated overview of the field and guidance for developing and evaluating more energy-efficient TinyML solutions.

AI 总结：WeightLoop是将实验反馈转化为预测器持续更新的框架，结合权重空间编程与递归自我改进循环，通过GPT-6 Astra生成参数编辑程序，可提升预测效用与发现效果。

链接：https://arxiv.org/abs/2611.07319

机构：Hong Kong Institute of Science and Innovation（香港科学与创新研究院）； University of Macau（澳门大学）； Sichuan University（四川大学）； Sun Yat-sen University（中山大学）； The University of British Columbia（不列颠哥伦比亚大学）； Westlake University（西湖大学）

作者：Mengran Li, Bo Li, Weijing Zhao, Chengyang Zhang, Boyu Zhang, Jinfeng Xu, Zelin Zang, Victor C. M. Leung, Stan Z. Li

英文摘要：Weight-space programming (WSP) shows that language models can reason about and rewrite neural network weights; recursive self-improvement (RSI) motivates feedback-driven improvement. Models guiding experiments receive concrete feedback: completed assays reveal prediction errors and inform what to test next. We introduce WeightLoop, a framework that turns experimental feedback into persistent updates to deployed predictors. GPT-6 Astra inspects a task specification and the model implementation, then writes a bounded, reusable parameter-editing program. As new measurements arrive, the program identifies prediction discrepancies, validates candidate edits using observed outcomes, and writes the selected changes back to the original model. The editor remains fixed while the predictor evolves, enabling repeated updates without further LLM calls or gradient retraining while preserving the original inference interface. In offline replay, targeted edits improve mean ranking utility in seven of eight endpoint-budget settings (all eight versus random edits) and mean final discovery across three sequential tasks. WeightLoop connects WSP with an RSI-inspired improvement loop, making experimental evidence a direct driver of model change and subsequent decisions.

AI 总结：本研究发现LLM微调中基于难度的数据选择方法效果不及随机选择，提出基于聚类的覆盖度优先选择方法，在单领域任务上与随机选择相当，在多领域Tülu 3数据集上优于所有方法。

链接：https://arxiv.org/abs/2611.07323

机构：University of Chicago（芝加哥大学）； The University of Texas at Austin（德克萨斯大学奥斯汀分校）； Allen Institute for Artificial Intelligence（艾伦人工智能研究所）

作者：Lalchand Pandia, Kanishka Misra, Allyson Ettinger

英文摘要：Selecting high-quality training data can reduce computation cost for LLM fine-tuning. Prior work on data selection developed a variety of difficulty-based scores, motivated by the intuition that harder examples are more informative. While effective for smaller classification models, we find that such methods often fail to outperform random selection for LLM fine-tuning. Our analysis reveals that random selection performs better because it is more representative of the underlying input space---i.e., has better coverage. To validate the importance of coverage, we implement a simple clustering-based selection method which selects data that is more representative of the underlying input distribution. On a set of diverse single-domain tasks, this method yields performance at or above that of random selection. On the diverse, multi-domain Tülu 3 mixture, where random sampling no longer guarantees a balanced subset, our method outperforms all methods. Overall, our results indicate that prioritizing coverage over difficulty leads to more effective data selection for LLM fine-tuning.

AI 总结：针对通用链接预测现有方法无法捕获高阶信息、领域适配性差的问题，提出AtlasULP框架，引入关系图谱并开发领域引导的提示式模型，在多领域图上性能优于现有SOTA方法。

链接：https://arxiv.org/abs/2611.07347

机构：Griffith University（格里菲斯大学）； La Trobe University（拉筹伯大学）； Tongji University（同济大学）

作者：Yujing Liu, Yixin Liu, Yu Zheng, Lianhua Chi, Alan Wee-Chung Liew, Heng Tao Shen, Shirui Pan

英文摘要： Link prediction (LP) is a widely applied task in graph learning. Conventional LP approaches typically follow a dataset-specific paradigm, requiring independent training for each graph and incurring high computational and maintenance costs in large-scale applications. Motivated by these limitations, recent work explores Universal Link Prediction (ULP), aiming to enable training-free inference on arbitrary unseen graphs. However, existing methods primarily rely on subgraph sampling strategies to construct universal representations, which are unable to capture higher-order information due to the exponential growth of higher-order neighborhoods. As a result, they often under-perform on graphs where long-range signals are essential, such as biological networks. Moreover, their prediction models are typically domain-agnostic and lack the ability to adapt to different graph domains, leading to limited generalization. To address these challenges, we propose an Atlas-based Universal Link Prediction framework (AtlasULP), which introduces the relation atlas to capture structural associations between node pairs from both global and local perspectives, enabling a comprehensive characterization of connectivity signals across diverse graph domains. Building upon this representation, we develop a domain-guided prompting ULP model, which generates domain-aware prompt tokens from contextual structures and performs prompt-guided in-context prediction for adaptive link inference. Extensive experiments demonstrate that AtlasULP consistently outperforms state-of-the-art methods across diverse graph domains.

AI 总结：针对测试时扩展中验证器不可用的问题，提出无需验证器的TTSP方法，在LiveCodeBench v6子集上，无验证器时mean@8绝对提升35%，且在各验证器强度下均优于基线。

链接：https://arxiv.org/abs/2611.07367

机构：ETH Zurich（苏黎世联邦理工学院）； Tufa Labs（图法实验室）

作者：Akira Yoshiyama, Akmal Ashirmatov, Jonas Hübotter, Ido Hakimi, Andreas Krause

英文摘要：Test-time scaling methods spend additional compute at deployment to improve a large language model's performance on a hard target problem. Most methods rely on a verifier that scores candidate solutions for correctness, using it to select among or refine the model's outputs. In many deployment regimes, however, a strong verifier on the target problem is unavailable. We introduce Test-Time Self-Play (TTSP), an approach to test-time scaling using asymmetric self-play without reliance on a verifier. On a difficulty-calibrated subset of LiveCodeBench v6, TTSP is the best-performing method at every verifier strength we evaluate, and the gap to the strongest baseline widens as the verifier weakens; in the no-verifier limit, TTSP yields a 35\% absolute improvement in mean@8. The gains hold across a range of verifier strengths, except when the base policy has near-zero coverage of the correct answer.

AI 总结：针对DPO中约束漂移问题，提出感知冲突的CoDPO方法，通过预对齐诊断量ρ*_eq降权冲突配对并添加自适应边际偏移，有效降低漂移且保留偏好学习，实现更优权衡。

链接：https://arxiv.org/abs/2611.07375

机构：University of Nevada, Reno（内华达大学雷诺分校）； Washington State University（华盛顿州立大学）

作者：Ping Liu, Yan Yan

英文摘要：Direct Preference Optimization (DPO) is increasingly used in settings where pairwise preferences coexist with pointwise criteria such as response length, answer correctness, or safety scores. When the two disagree, standard DPO treats pairs whose preferred output scores worse on the pointwise criterion (conflict pairs) in the same way as the remaining (matching) pairs. We characterize a systematic failure mode, constraint drift: for sufficiently small updates, conflict pairs tend to increase constraint cost while matching pairs tend to repair it, and the net change depends on their balance. This balance yields a closed-form pre-alignment diagnostic, $\rho^{*}_{\mathrm{eq}}$, computable from dataset statistics, that indicates whether a dataset is self-correcting or requires intervention. Building on this diagnostic, we propose Conflict-aware DPO (CoDPO), which down-weights conflict pairs with a strength set from $\rho^{*}_{\mathrm{eq}}$ rather than tuned, and adds an adaptive margin shift that addresses residual drift during training. Empirically, drift grows monotonically with conflict exposure on the conflict-dominated domains, and the intervention decision given by $\rho^{*}_{\mathrm{eq}}$ matches the observed drift on every domain, including one that it identifies as self-correcting. On all four domains of the main comparison, CoDPO reduces drift while largely preserving the preference learning of vanilla DPO. Compared with hard filtering and recent constrained and multi-objective baselines, CoDPO also achieves a more favorable drift-preference trade-off.

AI 总结：该研究针对预测多重性的对抗性场景，区分软硬操纵，发现灵活模型可通过硬操纵逆转个体预测并通过黑盒审计，还提出了相关监管与缓解建议。

链接：https://arxiv.org/abs/2611.07378

机构：Harvard University（哈佛大学）； Johns Hopkins University（约翰霍普金斯大学）； Cornell Tech（康奈尔科技学院）

作者：Cassandra C. Chou, Benjamin Q. Huynh, Allison Koenecke, Elizabeth T. Chin

英文摘要：Predictive multiplicity -- where similarly accurate models disagree on individual predictions -- can be leveraged to optimize secondary metrics (e.g., fairness) without loss in predictive performance. However, exploiting multiplicity opens the opportunity for motivated parties (or coding agents gaming their objectives) to steer predictions toward their own ends, such as in litigation or consumer disputes with automated decision-making tools. In these adversarial multiplicity settings, we distinguish soft manipulation, where parties optimize over innocuous model choices (e.g., hyperparameter tuning, feature engineering), from hard manipulation, where parties directly perturb model parameters. We find a sufficiently flexible model can be manipulated to reverse any individual prediction with minimal change in predictive performance while passing black-box audits, making hard manipulation viable in most regulatory settings. Soft manipulation has smaller effects but can pass white-box audits requiring full access to code and data. Even aggregate estimates can be manipulated: a model's estimated effect of climate change on wildfires can be fully reversed by hard edits and soft search. We find the Reconcile algorithm typically resolves most disagreements between models, but more often in favor of the manipulator when the opposing model is chosen adversarially compared to at random. We map which manipulations can plausibly pass existing regulations and disclosure policies, and recommend mitigations such as pre-registration, proof of model training, sensitivity analyses, multi-metric evaluations, and committing to versioned public models.

AI 总结：该研究针对多智能体编排中联合更新收益非各决策单独效应总和的问题，提出IAPO框架，在多类基准任务上以更少token达到最优准确率，移除交互项会显著降低代码任务准确率。

链接：https://arxiv.org/abs/2611.07495

机构：Emory University（埃默里大学）； Shanghai Jiao Tong University（上海交通大学）

作者：Kaixuan Liu, Guojun Xiong

英文摘要： Learned orchestration has emerged as a promising paradigm for large language model (LLM)-based multi-agent systems, in which a central orchestrator selects agents and controls the information they receive throughout a task. Despite rapid empirical progress, existing methods train the orchestrator by crediting each decision with its own advantage, implicitly assuming that the other decisions in the episode stay fixed. This assumption fails in orchestration, where the contribution of each decision to the final outcome depends on the decisions preceding it, so the gain of a joint update is not the sum of its per-decision effects. To address this challenge, we introduce \textbf{IAPO}, an interaction-aware policy optimization framework that explicitly accounts for interactions among decisions. Specifically, IAPO exactly decomposes the performance difference between two policies into the effects of individual decisions and their interactions, which we prove can account for half of the attainable gain, and then optimizes both from sampled trajectories while certifying that every update is an improvement. On long-input versions of GSM8K, MATH, DROP, MBPP and HumanEval, against fifteen published multi-agent and workflow-search systems on the same 7B executor, IAPO's deployed plans reach the strongest system's accuracy with fewer tokens, and removing the interaction term costs 7 to 17 points of accuracy on code.

AI 总结：本文针对交通网络约束学习提出物理约束的影响函数框架，揭示物理约束对训练样本影响的改变，其识别的有影响力观测值与经典方法存在差异，可用于交通领域数据相关任务。

链接：https://arxiv.org/abs/2611.07526

机构：University of California, Davis（加利福尼亚大学戴维斯分校）； University of Southern California（南加州大学）

作者：Hamidreza Attari, Yueyue Fan, Johannes Royset

英文摘要：Understanding how individual training observations influence model predictions is important for developing reliable and interpretable data-driven transportation systems. Existing influence-function methods, however, are primarily formulated for unconstrained learning and do not explicitly account for physical constraints that couple different components of a transportation network. Under such constraints, perturbing a training observation can affect model parameters and predictions only through physically feasible directions, potentially changing how its influence propagates across the network. This paper develops a hard physics-informed influence-function framework for constrained learning, in which training-data influence is characterized within the physically feasible space defined by network constraints. The analytical results reveal how physical constraints alter the magnitude, direction, and spatial distribution of training-sample influence relative to conventional influence measures. Using traffic prediction under flow-conservation constraints as a case study, we show that influential observations identified by the proposed framework differ systematically from those identified by classical influence functions. Through orthogonal decomposition, we quantitatively characterize how much of a perturbation in the training data is propagated through the network via the flow-conservation constraints and how much is removed by the constraint projection. The proposed influence framework provides a physically consistent approach for evaluating training-data influence in constrained transportation learning problems, with potential applications to data diagnostics, sensor evaluation, and targeted data collection.

AI 总结：MambaTG是基于Mamba的候选感知时序图框架，联合建模历史序列演化与候选-历史兼容性信号，在13个数据集上较13个基准实现了重复与未见边预测的高效且一致的性能。

链接：https://arxiv.org/abs/2611.07569

机构：Nanyang Technological University（南洋理工大学）； University of Montreal（蒙特利尔大学）； Mila – Quebec Artificial Intelligence Institute（米拉-魁北克人工智能研究所）

作者：Nguyen Hong Quan, Sitao Luan, Yilun Zheng, Lihui Chen

英文摘要：Future link prediction on temporal graphs involves both repeated interactions between previously connected node pairs and first-time interactions between pairs with no prior interaction. While repeated-edge prediction can exploit direct pairwise evidence, unseen-edge prediction must generalize beyond historical recurrence. Existing temporal graph models often exploit signals tied to recurring interactions or historical overlap, which are highly informative for repeated edges but become limited or unavailable for unseen edges, contributing to substantial performance degradation on unseen-edge prediction. Based on this observation, we identify two complementary signals for modeling sequential dynamics beyond historical recurrence: (1) historical-sequence evolution, which captures dependencies among the source node's chronologically ordered interactions, and (2) candidate--history compatibility, which captures how well a candidate fits the source node's evolving history. Based on the decomposition, we propose MambaTG, a candidate-aware Mamba-based framework that jointly models these two signals. Mamba provides a natural mechanism for this design: its selective recurrence captures the evolution of ordered interaction histories, while its input-dependent state transitions allow the history state to adapt to different candidate inputs. Extensive experiments on 13 temporal graph datasets against 13 representative baselines show that MambaTG achieves strong and consistent performance across both repeated and unseen edges while remaining computationally efficient.

AI 总结：针对硬件受限的LLM推理配置搜索难题，提出ACSeL框架，结合轻量级质量代理、少样本中位数校准与BandPareto算法，在两款硬件上验证其可提升部署可行性并降低RMQD差距。

链接：https://arxiv.org/abs/2611.07570

机构：University of Texas at Austin（德克萨斯大学奥斯汀分校）

作者：Bokun Wang, Hung-Yueh Chiang, Diana Marculescu

英文摘要： Serving large language models (LLMs) under tight memory and latency service-level objectives (SLOs), including on edge devices, requires jointly selecting the model size, batch size, quantization algorithm, and mixed-precision quantization recipes. Exhaustive search over this combinatorially large space is infeasible. Unlike data-center serving, which often prioritizes throughput or generation speed, resource-constrained deployments typically set memory and latency budgets as constraints and then seek the configuration with the highest generation quality among feasible ones. To this end, we propose ACSeL, a framework that identifies configurations predicted to satisfy deployment constraints while achieving high predicted generation quality. To enable efficient quality estimation, we use Relative Model Quality Degradation (RMQD), a proxy based on the divergence between the candidate and reference models' next-token distributions over a small calibration corpus. Across the evaluated model sizes and precisions, RMQD is inexpensive to evaluate and strongly correlated with downstream benchmark performance. ACSeL combines (i) a lightweight quality surrogate that predicts RMQD directly from a configuration; (ii) Few-shot Median Calibration (FMC), which calibrates operator-profile-based latency estimates using a small number of end-to-end profiles; and (iii) a structured search algorithm BandPareto for exploring layerwise mixed-precision configurations. Experiments on an NVIDIA RTX A6000 workstation and an NVIDIA Jetson Thor edge platform show that FMC substantially improves deployment feasibility, while BandPareto + FMC achieves the highest feasibility and the lowest RMQD gaps to the best measured configurations in the evaluation set.

AI 总结：该研究针对全序动作空间的对抗性在线学习问题，提出基于两点反馈的方法，解决分离单调与BV两类问题，在动态定价应用中验证其性能接近全信息率。

链接：https://arxiv.org/abs/2611.07629

机构：Meta

作者：Shiliang Zuo

英文摘要：We study adversarial online learning with expert recommendations in a totally ordered action space, where the experts' recommendations may change and be reordered arbitrarily. Our focus is two-point feedback: on each round the learner plays two current recommendations, observes both, and pays their average. We study two problems. In the separated monotone problem, the loss is the sum of a hidden nondecreasing component and a hidden nonincreasing component, and a query reveals the two components separately. In the BV problem, the losses are bounded and have bounded variation (BV) along the order, and a query reveals only the loss. In the separated monotone problem, two paid observations achieve \[ O\bigl(\log(|\cF|T)\sqrt{T\log|\cF|}\bigr) \] regret against the best of $|\cF|$ experts, within a logarithmic factor of the classical full-information bound. For the BV problem, we obtain $\wtO(\sqrt T+|\cF|)$ regret. We give an application in dynamic pricing with expert advice, showing that two-option menus matches the full information rate up to $\log$ factors. The main technical ingredient is a marginal-preserving endpoint flow estimator. Two observations reveal a loss increment along the order; random paths supply telescoping identities that turn this local information into unbiased loss estimates for all current recommendations. A reversible pair sampler preserves the learner's intended marginal at both played actions. The estimator's second-moment and stability bounds support online updates for both feedback models.

AI 总结：本研究针对LLMs的事后记忆问题，提出在冻结LLM主干与语言建模头间插入NVIB的即插即用框架，实验证实该方法可减少记忆提取且保持基准性能，为记忆缓解提供架构无关的实用方向。

链接：https://arxiv.org/abs/2611.07669

机构：Idiap Research Institute（伊迪亚普研究所）； EPFL（洛桑联邦理工学院）

作者：Dina EL Zein, Shashi Kumar, James Henderson

英文摘要：Recent studies have shown that large language models (LLMs) can memorize and reproduce copyrighted training data, raising serious privacy and intellectual property concerns. In this work, we investigate whether memorization can be mitigated without modifying or retraining the underlying model through an information bottleneck. We propose a modular plug-and-play framework that inserts a nonparametric variational information bottleneck (NVIB) between a frozen LLM backbone and its language modeling head. By compressing the final hidden representations before token prediction, the proposed bottleneck aims to preserve general linguistic knowledge while filtering out highly specific memorized sequences. We evaluate our approach on memorization extraction tasks involving book passages and on standard language modeling benchmarks to measure the trade-off between memorization reduction and downstream utility. Our experiments show that the NVIB layer substantially reduces extractability while maintaining competitive performance on benchmark tasks. These results suggest that adding an information bottleneck offers a practical and architecture agnostic direction for post-hoc memorization mitigation in pretrained LLMs.

AI 总结：本文提出校准目标噪声方案，仅通过聚合小批量梯度即可恢复曲率信息，适用于非线性模型与前馈ReLU网络，可加快优化并用于早停。

链接：https://arxiv.org/abs/2611.07671

机构：University of Helsinki（赫尔辛基大学）； KTH Royal Institute of Technology（瑞典皇家理工学院）

作者：Arash Jamshidi, Katsiaryna Haitsiukevich, Aristides Gionis, Kai Puolamäki

英文摘要：Estimating curvature information from gradients alone is a fundamental problem in machine learning. The challenge becomes particularly acute in settings where only aggregate mini-batch gradients are available, yet curvature information is still required for applications such as preconditioning in optimisation, sampling, and early stopping. Perhaps surprisingly, we show that a simple target-perturbation scheme suffices to recover such curvature information from gradients alone. Our method injects zero-mean noise into the targets, with variance calibrated to the local output curvature. The covariance of the resulting noisy batch gradients then recovers practical curvature matrices, without requiring per-sample gradients, second-order derivatives, or other internal network quantities. For a broad class of non-linear models, our scheme recovers the Generalised Gauss--Newton (GGN) matrix; for feed-forward ReLU networks, it additionally recovers the diagonal blocks of the population Hessian. We demonstrate that the resulting estimator is simple, lightweight, provably optimal, and easy to integrate into existing training pipelines. Experiments demonstrate accurate curvature recovery, faster optimisation when the estimator is used for preconditioning, and the utility of the estimated GGN for early stopping.

AI 总结：本研究针对连续时间随机过程开发工具变量框架，推导矩约束与估计方法，扩展至非线性因果效应，在混杂连续时间系统中表现优于现有方法。

链接：https://arxiv.org/abs/2611.07674

机构：Technical University of Munich（慕尼黑工业大学）； Helmholtz Munich（亥姆霍兹慕尼黑中心）； Munich Center for Machine Learning（慕尼黑机器学习中心）； Center for Statistics, Department of Finance（统计中心（金融系））； Copenhagen Business School（哥本哈根商学院）

作者：Lars Pennig, Søren Wengel Mogensen, Niki Kilbertus

英文摘要： Causal questions often arise in systems that evolve in continuous time. While instrumental variable (IV) methods provide a principled approach to identify causal effects in static or discrete-time settings, even in the presence of unobserved confounding, they do not yet extend to continuous-time dynamics. In this work, we develop an IV framework for stationary continuous-time stochastic processes parameterized by stochastic differential equations (SDE), where the instrument itself takes the form of a stochastic process. We encode the structural sparsity and causal effects directly at the infinitesimal level of the drift, and contrast these assumptions with the IV assumptions on the discretized version of the same process. Based on these structural assumptions, we derive moment restrictions using path-valued instruments. We show how these restrictions simplify for Ornstein-Uhlenbeck processes and present an identification condition together with a corresponding consistent estimation method. Finally, we extend our estimation framework to nonlinear causal effects by expressing the path-based moment restrictions leveraging the signature kernel. Empirically, our method recovers causal effects in confounded continuous-time systems where maximum-likelihood-based estimation and discrete-time IV methods are systematically biased.

AI 总结：提出基于简化D-Vine Copula的非参数互信息估计器VINE，在中低维数据集上，其相比SotA神经互信息估计器，计算成本更低且精度有竞争力。

链接：https://arxiv.org/abs/2611.07693

机构：University of Oslo（奥斯陆大学）； Integreat – Norwegian Centre for Knowledge-driven Machine Learning（Integreat——挪威知识驱动机器学习中心）； TRUST – The Norwegian Centre for Trustworthy AI（TRUST——挪威可信人工智能中心）； University of Cambridge（剑桥大学）； MRC Biostatistics Unit（医学研究理事会生物统计学部）

作者：Sigurd Holmsen, Berit Omli Øksnes, Ingrid Hobæk Haff, Sylvia Richardson, Ali Ramezani-Kebrya

英文摘要：We introduce VINE: a non-parametric mutual information (MI) estimator based on simplified D-vine copulas. Approximating the joint density with a simplified D-vine allows the density ratio to simplify to a product of pair copulas, which we can estimate non-parametrically. This structural simplification yields VINE, a flexible and computationally efficient MI estimator. We compare VINE against SotA neural MI estimators on a range of datasets, both under settings where the simplified D-vine assumption holds and where it is violated, demonstrating competitive accuracy at a lower computational cost for low- to medium-dimensional datasets.

AI 总结：该研究针对在线策略蒸馏的反馈幅度差异问题，提出SD-OPD方法，以学生预训练基础为参考调整更新系数，在Qwen3模型上使数学推理任务性能较OPD提升7.76至10.37个百分点。

链接：https://arxiv.org/abs/2611.07722

机构：Zhejiang University（浙江大学）； Ant Group（蚂蚁集团）

作者：Bozhen Fang, Linhao Zhong, Xuehao Wang, Huanyi Zheng, Hao Chen, Chunhua Shen

英文摘要：On-policy distillation (OPD) improves language-model reasoning through dense teacher feedback on student-generated this http URL first observe an amplitude disparity in this feedback: on math reasoning responses, style-related positions exhibit 3.5 times the mean absolute teacher--student log-probability difference of mathematical this http URL OPD updates by these discrepancies thus directs much of the update strength toward matching the teacher's expression. We instead take the student's own pretrained base as a reference: the student--base difference records the token preferences acquired during post-training and provides supervision that does not depend on the student's distance from the teacher. Building on this reference, we propose Student-Delta On-Policy Distillation (SD-OPD), which uses centered student--base differences as update coefficients, keeps each coefficient only when its sign agrees with centered teacher--student feedback, and recomputes the student signal every training round. Across 5 teacher--student configurations with Qwen3 students of 1.7B to 8B parameters, SD-OPD improves six-task mathematics average pass@1 (avg@16) over OPD by 7.76 to 10.37 percentage points. Code is available at this https URL.

AI 总结：本文通过分析不同尺度因子归一化空间中分类与特征学习的损失函数，发现适中尺度因子可同时实现可靠分类与良好特征结构，且BCE损失在长尾和开放集识别中更具优势。

链接：https://arxiv.org/abs/2611.07741

机构：School of Artificial Intelligence, Shenzhen University（深圳大学人工智能学院）； National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University（深圳大学国家大数据系统计算技术工程实验室）； Guangdong Provincial Key Laboratory of Intelligent Information Processing, Shenzhen University（深圳大学广东省智能信息处理重点实验室）； College of Computer Science and Software Engineering, Shenzhen University（深圳大学计算机与软件学院）； Big Data Institute, University of Oxford（牛津大学大数据研究所）； Institute of Image Understanding Research, North Minzu University（北方民族大学图像理解研究所）

作者：Qiufu Li, Huibin Xiao, Linlin Shen, Weijia Fan, Xi Jia, Yongdong Huang

英文摘要：Classification and feature learning are two closely interrelated fundamental tasks. Feature learning in supervised settings is typically achieved by training classification models, under the common assumption that higher classification accuracy implies better feature quality. In this paper, we first point out that classification accuracy reflects the sample-wise local properties of a model on a dataset, whereas feature learning aims to learn features with good dataset-level global properties, such as intra-class compactness and inter-class separability. From the first principle of feature learning, we redesign the binary cross-entropy (BCE) loss, a counterpart of the cross-entropy (CE) loss used in classification. Then, in normalized feature space, we analyze the minima of the BCE and CE losses and show that both can lead to neural collapse, where sample features from different classes become maximally compact and maximally separated. More importantly, by investigating the convergence behaviors of the two losses as the scale factor $\gamma$ varies in the normalized space, we reveal how classification accuracy and feature quality change differently across spaces of different sizes. Under a fixed training strategy, a very large $\gamma$ achieves perfect classification accuracy but slows convergence toward the ideal feature geometry, whereas a very small $\gamma$ makes classification highly sensitive to small perturbations. A moderate $\gamma$ is therefore necessary for simultaneously achieving reliable classification and well-structured features. Extensive experiments validate these findings and demonstrate that BCE loss exhibits advantages in long-tail and open-set recognitions, as it better aligns with the global requirements of feature learning.

AI 总结：本文提出QS-TCN方法，结合波动率记忆统计估计与时间卷积学习，在多元时间序列排序任务中，于合成系统及Finance、fMRI基准上取得优于部分对比方法的排序效果。

链接：https://arxiv.org/abs/2611.07746

机构：Hassan II University（哈桑二世大学）； IRIT（IRIT研究所）； Université Toulouse Capitole（图卢兹第一大学）； EMSI（EMSI学院）

作者：Kaoutar El Azaar, Mohamed Hamlich, Moncef Garouani, Franck Ravat, Asmae Chakir, Mohamed Tabaa

英文摘要：Volatility dynamics can carry information about temporal relationships that is not apparent in the conditional mean. Extracting this information is difficult because the persistence of each variable can obscure its dependence on other variables. This paper presents QS-TCN, a method for ordering multivariate time series that combines statistical estimates of volatility memory with temporal convolutional learning. Persistence estimates obtained by quasi-maximum likelihood and temporal modes extracted by the Matrix Pencil method define structured filters within the network. Together with learned filters, they represent the history of the target and support comparisons of directional evidence between variables. These comparisons are combined into a global order of precedence, which does not identify a causal graph. The method is evaluated on synthetic systems with linear and nonlinear volatility interactions and on the Finance and fMRI benchmarks, whose reference relations are known. Across the four synthetic volatility regimes studied, QS-TCN reaches a higher mean ordering quality than an external lead--lag ordering and its squared-input adaptation, although the gain over the adaptation is uncertain in two regimes. Applications to currency, gold and cryptocurrency markets are reported descriptively.

AI 总结：该研究提出ConvPE卷积位置嵌入插件，通过轻量因果卷积改进Transformer模型，兼容多种现有组件，在多任务中性能优于多个基线。

链接：https://arxiv.org/abs/2611.07749

机构：Nanjing University（南京大学）

作者：Zehao Wang

英文摘要：ConvPE (Convolutional Position Embedding) improves both the performance and efficiency of Transformer models in language modeling, long-context extension, and vision. It injects a windowed, content-dependent bias into attention scores through a lightweight causal convolution over a low-dimensional projection of the hidden states. As a lightweight plug-in, ConvPE is compatible with RoPE, attention masks, and KV concatenation. It also supports high-dimensional position embeddings via high-dimensional convolutions. In a reproduction, ConvPE outperforms five 1D baselines and eight 2D baselines. The code is publicly available at this https URL.

AI 总结：该研究针对SAE训练的激活存储瓶颈，提出有限数据的缩放定律，验证其在多模型、多SAE类型上的有效性，给出减少激活存储8--64倍且质量损失可忽略的实用训练规则。

链接：https://arxiv.org/abs/2611.07756

机构：Yandex； T-Tech

作者：Nikita Koriagin, Nikita Balagansky, Daniil Gavrilov

英文摘要：Activation storage is often a bottleneck in sparse autoencoder (SAE) training. We study how many unique activations an SAE needs when its total training budget is fixed. Quality improves rapidly as the buffer grows, then reaches a plateau where replaying the buffer is nearly as effective as using fresh activations throughout training. Beyond this point, additional unique activations bring little benefit to reconstruction or interpretability, so the buffer need not grow with the training budget. We describe this tradeoff between diversity and repetition with a scaling law for limited data. We validate it across dictionary sizes and token budgets, early, middle, and late layers of Llama-3.1, Qwen3 model sizes, and BatchTopK, TopK, and JumpReLU SAEs. Sweeps using mathematical text and Python code test the effect of changing the activation source. Evaluation covers explained variance (EV), reconstruction MSE, feature absorption, automated interpretability, and sparse probing. These results give a practical training rule: choose the smallest buffer that reaches the quality plateau and replay it for the remaining budget. In our experiments, this reduces activation storage by $8\text{--}64\times$ with negligible quality loss.

AI 总结：针对带任意延迟的零和博弈学习问题，提出MAO-OMD算法，可同时在自我博弈中获最优快速率、对抗任意对手时保持极小极大最优保证，且该最优无法通过标准黑盒降维实现，还可扩展至多玩家及玩家依赖延迟场景。

链接：https://arxiv.org/abs/2611.07773

机构：CREST, ENSAE, IP Paris（CREST、法国国立统计与经济管理学院、巴黎IP大学）； Universitat Pompeu Fabra（庞培法布拉大学）； National University of Singapore（新加坡国立大学）； University of Toronto, Vector Institute（多伦多大学向量研究所）； University of Iowa（爱荷华大学）

作者：Emmanuel Esposito, Andrew Jacobsen, Hao Qiu, Alexander Ryabchenko, Mengxiao Zhang

英文摘要：We study learning in zero-sum games with delayed feedback. We introduce Multi-Anchor Optimistic Online Mirror Descent (MAO-OMD), a single algorithm which obtains optimal fast rates in self-play on smooth games, while simultaneously maintaining minimax optimal guarantees against arbitrary opponents. In particular, in smooth zero-sum games with arbitrary and unknown feedback delays, MAO-OMD obtains $O(\sigma_{\max}+1)$ social and individual regrets when used by both players, where $\sigma_{\max}$ denotes the maximum number of outstanding gradients, while also guaranteeing minimax optimal $O(\sqrt{T+d_{\mathrm{tot}}})$ individual regret against arbitrary opponents, where $d_{\mathrm{tot}}$ is the total delay. Moreover, we show that these optimal rates cannot be obtained simultaneously using the standard black-box reduction to the undelayed setting. We also extend our results to multi-player games, and to games with player-dependent delay sequences.

AI 总结：本研究在伯努利推理任务中发现，语言模型的匹配预测不代表信念更新等价，证据格式、获取路径会影响信念动态，重现即时预测的干预在后续证据下效果大幅下降。

链接：https://arxiv.org/abs/2611.07913

机构：The University of Tokyo（东京大学）； RWTH Aachen University（亚琛工业大学）； Harbin Institute of Technology, Shenzhen（哈尔滨工业大学深圳校区）； The Hong Kong University of Science and Technology（香港科技大学）

作者：Weihan Li, Junhao Wu, Xinlei Chen, Tianshi Zheng

英文摘要：Language models can produce similar probabilistic predictions after observing outcomes, reading summaries, receiving instructions, or undergoing activation interventions. Whether these routes induce equivalent belief states remains unclear: agreement in current predictions does not establish how a model will respond to subsequent evidence. We investigate this distinction in a Bernoulli inference task with an analytic Bayesian reference. Across three open-weight language models, we prepare states through four textual routes and two activation-level interventions, measure both next-outcome continuations and explicit probabilistic estimates, and append identical new evidence from each prepared state. We quantify the persistence of the initial belief through its effective evidence weight, the number of new observations it is worth. We find that the format of evidence decides which readouts express its content: outcome sequences primarily affect continuations, whereas summaries and stated rates primarily affect explicit estimates. Moreover, effective weights differ substantially across routes, with summaries receiving weights that scale weakly with their stated observation counts and elicitation questions changing inferred weights by orders of magnitude. In a one-shot steering setting, an intervention that reproduces a natural record's immediate prediction loses most of its effect under subsequent evidence. These results show that agreement in current readouts is insufficient to establish equivalence in belief updating: how a belief is acquired and how it is read jointly determine its subsequent dynamics.

AI 总结：提出CoRD耦合回滚蒸馏方法，通过最大耦合抽取回滚数据填补策略蒸馏的状态监督盲区，在6个Qwen系列学生模型的5个数学基准上提升了pass@1和pass@8指标。

链接：https://arxiv.org/abs/2611.07959

机构：Tsinghua University（清华大学）； Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））； Nanjing University（南京大学）； Huawei（华为）

作者：Guhan Chen, Bohan Li, Jun Rao, Hejin Wang, Songtao Tian, Yexin Xie, Xiaojun Meng, Peng Yuan, Jiansheng Wei, Xuebo Liu, Zixiong Yu

英文摘要：On-policy distillation trains a student on its own rollouts, so states the teacher would reach but the student does not are seldom supervised. We introduce CoRD (coupled-rollout distillation), which keeps the on-policy loss and moves only the distribution of training states toward the teacher, drawing rollouts from a geometric interpolation of student and teacher through the maximal coupling of speculative decoding. The coupling keeps the student's token wherever the two distributions agree and otherwise substitutes a token from the part of the target the student underweights, so interventions are as rare as any exact sampler allows and recede as the student learns. Across six base students from two model families (Qwen3 and Qwen3.5, 0.6B to 4B), CoRD raises average pass@1 over on-policy distillation by 0.9 to 2.0 points and pass@8 by 1.2 to 3.4 points on five public math benchmarks at equal steps and data. On frozen teacher-reached states, a 4B student trained with CoRD reaches in 16 steps a gap closure that on-policy distillation does not reach in 80, and substitution rules intervening at least as often in other directions fall short of CoRD.

AI 总结：本文针对CPWL KAN，给出了其最大线性区域数的匹配上下界，改进了已有上界，还证明T≥2时其区域数相对ReLU网络有多项式优势，为KAN架构设计提供了基础。

链接：https://arxiv.org/abs/2611.07973

机构：Shenzhen Institutes of Advanced Technology（深圳先进技术研究院）； Chinese Academy of Sciences（中国科学院）； Gaoling School of Artificial Intelligence（高瓴人工智能学院）； Renmin University of China（中国人民大学）

作者：Ye Su, Yong Liu

英文摘要：Schoots et al.~\citep{schoots2025relating} (AISTATS 2025) established a generally loose upper bound on the number of linear regions of continuous piecewise-linear (CPWL) Kolmogorov--Arnold networks (KANs), without providing a lower bound. We consider a $T$-layer network with width vector $\mathbf n=(n_0,\ldots,n_T)$, input dimension $n_0=d$, output width $n_T=1$, and at most $k\geq2$ affine segments per edge. Our matching upper and lower bounds determine the tight growth order $\Theta_{\mathbf n}(k^{\sum_{\ell=1}^{T}\rho_{\ell-1}})$ in $k$, where $\rho_{\ell-1}:=\min_{0\leq s\leq\ell-1}n_s$. When every hidden width is at least $d$, the tight order simplifies to $\Theta_{\mathbf n}(k^{dT})$. The upper bound of \citep{schoots2025relating} is $k^{\sum_{\ell=1}^{T}n_{\ell-1}n_\ell}$. Our upper bound improves it by a factor of $\Theta_{\mathbf n}\!\left(k^{\sum_{\ell=1}^{T}n_{\ell-1}n_\ell-\sum_{\ell=1}^{T}\rho_{\ell-1}} \right)$, with strict asymptotic improvement whenever a hidden layer has width greater than one. We further compare CPWL KANs with rectified linear unit (ReLU) networks using dense parameterizations with at most $\Lambda$ parameters per model and the same fixed nonlinear depth $T$. Counting trainable knot locations and optimizing the widths and $k$, we prove that the maximum region count of CPWL KANs divided by that of ReLU networks is $\Theta_{d,T}(\Lambda^{d\lfloor T/2\rfloor})$ as $\Lambda\to\infty$, yielding a polynomial advantage for KANs when $T\geq2$. These tight bounds characterize the attainable geometric complexity of CPWL KANs, providing a foundation for understanding their expressive power and guiding architectural design.

AI 总结：提出QRAE方法，通过查询相关注意力熵对DeepSearch智能体的检索块进行评分筛选，在不修改注意力算子的情况下，提升任务成功率并降低延迟与token消耗。

链接：https://arxiv.org/abs/2611.07976

机构：Tsinghua University（清华大学）； Huawei Technologies Ltd.（华为技术有限公司）

作者：Yingjie Tan, Haiquan Qiu, Quanming Yao, Manyi Zhong, Xianzhi Yu, Yaqing Wang

英文摘要：DeepSearch LLM agents repeatedly search and carry retrieved sources into subsequent reasoning steps. This expanding context increases token consumption and latency, although many results do not contribute to later reasoning. We propose Query-Relevant Attention Entropy (QRAE), an online context compression method that uses sparse evidence-assessment heads to decide which intact retrieved chunks should persist across reasoning steps. It scores bounded candidate batches through paired prefills: a query-free control suppresses shared attention responses, while normalized entropy measures the breadth of the positive response within each chunk. High-scoring chunks enter a persistent evidence whitelist and are included in the reasoning context. Uncertain chunks remain outside the reasoning context but are retained in a candidate cache, where they can be retrieved for later reevaluation. QRAE requires neither an additional generative compression pass nor modification of the attention operator. Across multiple DeepSearch benchmarks and reasoning models, QRAE improves task success by an average of 2.35 percentage points and reduces end-to-end latency and reported token usage by an average of 30% and 45%, respectively. These benefits generalize to a model with nearly 100B parameters.

AI 总结：本文研究具有$\ell^p$系数范数的再生核（拟）巴拿赫空间理论，针对非凸正则化框架，构造相关空间并建立插值与正则化学习的存在性、稀疏性及表示定理等结果，揭示不同正则化 regime 的差异并通过数值实验验证。

链接：https://arxiv.org/abs/2611.08032

机构：School of Mathematics, Sun Yat-sen University（中山大学数学学院）； Department of Mathematics and Statistics, Old Dominion University（奥多明尼昂大学数学与统计系）

作者：Tianfang Wang, Guohui Song, Haizhang Zhang

英文摘要：This paper advances the theory of reproducing kernel Banach spaces (RKBSs) by investigating the general $\ell^p$ regularization framework, with a particular focus on the non-convex regime $0

1$, we construct RKBSs using a fixed countable dictionary, establish minimizer existence, and identify the optimal relaxed representer factor through an off-sample coefficient operator. These results reveal fundamental differences between the regimes $0

1$. Numerical experiments compare the accuracy and sparsity of solutions for these three cases.

AI 总结：该研究提出代理证书框架，将鲁棒性证书委托给廉价代理，降低随机平滑的高成本，可作为独立证书或回退机制，为大模型提供高效鲁棒性保证。

链接：https://arxiv.org/abs/2611.08035

机构：CISPA Helmholtz Center for Information Security（CISPA亥姆霍兹信息安全中心）； University of Cologne（科隆大学）

作者：Soroush H. Zargarbashi, Mohammad Sadegh Akhondzadeh, Aleksandar Bojchevski

英文摘要：Modern classifiers remain sensitive to small input perturbations, while strong model-agnostic robustness certificates such as randomized smoothing are often too expensive for deployment, requiring an extensive number (e.g. 2000) of model forward passes per input. We propose a framework that delegates certification to a cheap proxy function while statistically controlling its deviation from the original sound certificate. Given an exchangeable unlabeled calibration set, we tune the proxy via robust conformal risk control to guarantee a bounded marginal (rather than pointwise) false-approval rate: the probability, over future inputs, that the proxy accepts an input whose prediction is flipped due to perturbation. The resulting proxy certificate can be used either as a standalone marginal certificate or as a fallback mechanism that invokes the expensive certificate only when the proxy rejects. We provide both single-sample and sample-heavy calibration recipes. Empirically, as a standalone certificate our method reduces the test-time certification cost to a single model forward while retaining much of the certified performance of randomized smoothing at small radii, and as a fallback it recovers the certified ratio of randomized smoothing at a more modest speedup. This enables cheap yet effective robustness guarantees for larger architectures such as vision transformers and language models.

AI 总结：本研究通过15个受控场景阐释机器学习性能指标的影响因素，整理22个混淆矩阵指标，验证相关检验，提供指标表与报告清单，为性能指标使用提供场景化指南。

链接：https://arxiv.org/abs/2611.08048

作者：Mehmet Baygin, Sengul Dogan, Turker Tuncer

英文摘要：Performance metrics can yield different conclusions from the same predictions. This methodological tutorial explains the effects of class prevalence, decision thresholds, aggregation rules and software conventions through controlled scenarios. Twenty-two confusion-matrix metrics are organized in a prevalence-coded map. Fifteen scenarios cover binary and multi-class classification, score discrimination, probability calibration, regression, uncertainty and consistency checks. Inputs are fixed in released files, and one factor is varied in each scenario. All 205 verification checks passed; 202 were independent and 3 were internal. An always-negative rule achieved an accuracy of 0.990 at a prevalence of 0.010, with balanced accuracy of 0.500. At fixed sensitivity and specificity, precision fell from 0.889 to 0.008 as prevalence fell from 0.5 to 0.001. Three probability variants shared a ROC-AUC of 0.825 but had calibration slopes of 2.08, 1.04 and 0.52. With 10 participants, participant-majority accuracy of 0.700 had a 95% interval of 0.348-0.933. Fold-averaged scores were falsely flagged as inconsistent under an inappropriate pooled-test-set model. A goal-based metric table and a reporting checklist are provided. The scenarios illustrate mechanisms and do not estimate their frequency in published studies. A literature audit and a real-data demonstration are specified as future protocols and were not executed.

AI 总结：本研究测试Claude Opus 5.5对神经网络权重的读取能力，发现其为局部算术加数据集先验而非权重流推理，还提供了审计权重空间主张的协议。

链接：https://arxiv.org/abs/2611.08070

机构：BRAC University（BRAC大学）

作者：Nazmus Sakib Ahmed

英文摘要：Large language models (LLMs) have recently been reported to read and write the raw weights of neural networks directly, a capability termed weight-space programming. We ask whether the reading side of this capability reflects reasoning about weight flows or a combination of local arithmetic at the network's edges and memorized priors over public datasets. We test Claude Opus 5.5 on small convolutional networks of about 125,000 parameters trained on five image datasets, providing every weight as plain text with no tools, no forward passes, and no dataset or class names. The model performs accurate local calculations: batch-normalization statistics lead it to infer about three optimizer steps on networks stopped after three updates, and output-layer sign patterns recover the vehicle/animal split of CIFAR-10. The sessions report reading the first and last layers closely and sampling the middle, and several claims built on partial tracing fail verification, although we did not run a direct forward-pass test. Semantic judgments lean on familiar datasets: the model names CIFAR-10 for CIFAR-10 networks but also for all three fast-food networks, relabels KMNIST as MNIST, reads a 99.57%-accurate SAT-6 network as CIFAR-10 animal photos, and calls "frog" the output favored by a feature that fires most on trees. Permuting CIFAR-10's labels preserves the output split at the new indices, but every class pair the model interprets is then wrong, indicating reliance on remembered label order. These are exploratory results from mostly single runs. We conclude that zero-shot weight reading by this model is better described as local arithmetic plus dataset priors than as weight-flow tracing, and we provide a protocol for auditing future weight-space claims.

AI 总结：本文针对两类移动成本下满足二次增长的凸函数追逐问题，提出CPD算法并建立R-OBD的紧分析，得到最优无维度竞争比，填补了相关问题的理论空白。

链接：https://arxiv.org/abs/2611.08078

机构：Nanjing University（南京大学）； Nanjing University of Science and Technology（南京理工大学）

作者：Linzhe He, Sifan Yang, Wei Jiang, Lijun Zhang

英文摘要：Chasing convex function has attracted much attention in recent years, which can be formulated as a game between a learner and an adversary. In each round, the adversary reveals a convex function, and the learner selects a decision and incurs a hitting cost and a movement cost for changing its decision. The performance of the learner is measured by the \emph{competitive ratio}, which compares its cumulative cost with that of an optimal offline decision sequence. For Euclidean movement, existing upper and lower bounds for $\kappa$-well-conditioned functions exhibit a polynomial gap. For squared Euclidean movement, dimension-free competitive guarantees have been established under quadratic growth, but the optimal competitive ratio remains unclear. In this paper, we establish optimal dimension-free competitive ratios for convex function chasing under two types of movement costs. For Euclidean movement, we propose Clipped Proximal Descent (CPD), which combines a balanced proximal step with a clipped correction toward the current minimizer. CPD achieves an $O(\kappa^{1/3})$ competitive ratio for smooth convex functions satisfying quadratic growth, improving the previous $O(\sqrt{\kappa})$ upper bound and matching the existing $\Omega(\kappa^{1/3})$ lower bound. For squared Euclidean movement, we establish the exact optimal dimension-free competitive ratio for convex functions satisfying quadratic growth. Our result is based on a new tight competitive analysis of Regularized Online Balanced Descent (R-OBD) and a matching lower bound against randomized algorithms.

AI 总结：提出E-Gaborformer，一种结合Gated DeltaNet-2循环层与滑动窗口注意力的严格因果语音识别器，在LibriSpeech等实验中表现更优，训练速度更快，且权重可兼容不同解码方式。

链接：https://arxiv.org/abs/2611.08092

机构：Deepdub

作者：Nir Krakowski, Tamir Cohen, Ohad Zohar, Chen Zarfati

英文摘要：Real-time voice agents need a speech recognizer that never waits for future audio and whose memory per call does not grow with the length of the call. We present E-Gaborformer, a strictly causal encoder that combines Gated DeltaNet-2 recurrent layers with sliding-window attention in a 3:1 ratio, and builds every temporal convolution from a fixed dictionary of causal Gabor atoms. Its streaming state is constant, about 19 MB per call at production scale including the prediction network, where a causal-attention encoder of the same size accumulates over 1 GB after ten minutes. In a controlled causal study on LibriSpeech, E-Gaborformer leaves the initial plateau earlier, reaches a given accuracy up to 18% sooner, and matches or beats an E-Branchformer trained with the identical front end and recipe. At production scale, a 0.6B-parameter Hebrew and English streaming transducer built on E-Gaborformer has 28 to 31% lower word error rate (WER) than NVIDIA's Nemotron 3.5 ASR Streaming, a model of the same size, on English voice-agent traffic, and up to 20% lower Hebrew WER than our previous production system. The same weights serve strictly causal and 480 ms-lookahead decoding with no loss in accuracy, and a two-class formulation of the RNN-T lattice trains the transducer 2.6x faster with exact loss and gradients.

AI 总结：本研究针对张量网络核机的秩选择难题，提出梯度提升框架GB-CPKM，通过依次添加固定秩学习器构建预测器，相比传统交叉验证秩选择方法，训练速度最高提升6倍、内存最多减少2倍，且预测性能相当。

链接：https://arxiv.org/abs/2611.08093

机构：Delft University of Technology（代尔夫特理工大学）； Delft Center for Systems and Control（代尔夫特系统与控制中心）

作者：Albert Saiapin, Kim Batselier

英文摘要： Tensor network kernel machines (TNKMs) combine nonlinear features with low-rank tensor representations to model complex relationships efficiently. However, selecting an appropriate tensor rank remains challenging, as conventional rank-search methods require training multiple candidate models, increasing computational costs. In this work, we focus on a canonical polyadic (CP) tensor network parameterization and propose GB-CPKM, a gradient boosting framework that builds a higher-capacity predictor by sequentially adding fixed-rank learners. This approach provides a new perspective on rank selection by incrementally increasing model capacity, with ensemble size selected based on validation performance rather than through explicit rank search. Experiments on real-world regression benchmarks show that GB-CPKM trains up to 6x faster and uses up to 2x less peak memory than a conventional rank-selection method based on cross-validation, while achieving comparable predictive performance. These results highlight the potential of gradient boosting to improve the computational efficiency and scalability of tensor-network learning.

AI 总结：本研究对比仅用卫星观测与结合ERA5再分析数据的机器学习模型，发现后者在天气锋面检测的临界成功指数上提升0.173，且低层风对检测性能影响最大。

链接：https://arxiv.org/abs/2611.08148

机构：University of Amsterdam（阿姆斯特丹大学）； EUMETSAT（欧洲气象卫星开发组织）

作者：Ioannis Dravilas, Miruna Stoicescu, Roope Tervo, Carlos Horn, Oriol Hinojo Comellas, Sina Montazeri, Andrea Meraner, Ana Lucic

英文摘要：Weather fronts are transition regions between air masses with contrasting thermodynamic properties and are associated with mid-latitude precipitation and severe weather. We investigate machine learning (ML) methods for detecting weather fronts and quantify the contribution of reanalysis data to performance. We compare two configurations of the same ML model for pixel-wise frontal classification over the North Atlantic and Europe: (1) using only geostationary infrared satellite observations, and (2) additionally using 28 ERA5 reanalysis fields. The satellite-only model achieves a neighbourhood critical success index at a 150 km tolerance (CSI@150) of $0.402$, which lies within the range of agreement between analyses from different meteorological services. The satellite$+$ERA5 model achieves $0.575$, an improvement of $0.173$. Our analysis shows that low-level winds are the least predictable from the satellite representation ($R^2 = 0.23$, compared with $0.88$ for temperature), while removing winds or moisture causes the largest decreases in detection performance.

AI 总结：该研究发现排列对齐神经网络线性插值时存在BUMP现象，提出权重预测混合集成方法，其测试准确率优于标准logit集成且校准权衡更优。

链接：https://arxiv.org/abs/2611.08182

机构：HUN-REN Alfréd Rényi Institute of Mathematics（HUN-REN阿尔弗雷德·莱尼数学研究所）； Eötvös Loránd University（厄特沃什·罗兰大学）

作者：Melinda Flóra Kiss, Márton Muntag, Adrián Csiszárik, Dániel Varga

英文摘要：We identify an intriguing phenomenon when linearly interpolating between permutation-aligned neural networks: test accuracy briefly improves near the interpolation endpoints before dropping toward the midpoint, which we term BUMP (Brief Unexpected Mid-interpolation Peak). We observe that BUMP appears consistently across datasets, architectures, training regimes, and remains robust under substantial perturbations such as early checkpoints and extreme pruning. Moreover, controlled ablations suggest that it cannot be attributed to trivial regularization effects. While similar near-endpoint gains are visible in prior interpolation plots, they have not been explicitly documented or systematically and thoroughly analyzed. Leveraging this effect, we propose a simple, novel ensembling scheme: weight-prediction mixed ensembling that combines weight mixing with prediction-space mixing. Across our experiments, the resulting ensembling method consistently outperforms standard logit ensembling in test accuracy and yields improved calibration trade-off, highlighting complementary benefits of weight- and prediction-space mixing.

AI 总结：本研究评估三个sEMG基础模型的跨被试泛化能力，发现其仅在会话内场景提升性能，跨被试场景优势消失，少被试数据集上经典ML基线性能更优。

链接：https://arxiv.org/abs/2611.08199

机构：Imperial College London（帝国理工学院）

作者：William Raftery, Dario Farina

英文摘要：sEMG Foundation Models aim to learn general purpose representations over diverse subjects and recording conditions that boost performance on a variety of downstream tasks. However, existing evaluations make it difficult to discern whether pretraining alone is sufficient to combat the cross-subject overfitting which makes machine learning on sEMG so challenging. We conduct a systematic evaluation of three sEMG foundation models across three gesture classification datasets under both intra-session and inter-subject splits, using three fine-tuning strategies designed to isolate the contribution of pretraining. We find that pretraining boosts performance when train and test samples are drawn from the same recording session, but this benefit largely disappears under inter-subject evaluation. In low subject count datasets, pretraining provides little to no benefit and classical ML baselines match or exceed foundation model performance.

AI 总结：该研究针对扩散对接中几何相关性不足的问题，以DiffDock为对象，对比广泛延续、聚焦扭转监督等方法，发现聚焦监督未在固定预算下提升对接成功率。

链接：https://arxiv.org/abs/2611.08249

机构：University of Central Florida（中佛罗里达大学）

作者：Michael Brown, Maryam Rahimimovassagh, Mehdi Yazdani-Jahromi, Sanjeda Sara Jennifer, Ivan Garibay

英文摘要： Diffusion-based molecular docking requires learning a corrective score field across a broad corruption distribution, yet individual pose errors can differ substantially in their structural consequences. This creates a post-training conflict between preserving broad score-field coverage and concentrating supervision on corrections that appear more relevant to docking geometry. Using DiffDock, a fixed-budget study compares broad continuation with torsional supervision focused on Cartesian consequence or conditional target consistency, and with a broad/focused mixture. Evaluation separates fixed score fidelity, generated-pose quality, best-of-10 sampling capacity, and confidence-selected Top-1 docking performance. At <5 Angstrom, broad continuation had the highest mean Top-1 success among trained policies, and focused or mixed supervision showed no consistent advantage under the tested budget. The starting model had better fixed score fidelity, whereas the evaluated broad-continuation model had higher mean Top-1 success with the same aggregate best-of-10 recovery. Focusing supervision on structurally consequential or more consistently specified corrections did not consistently improve returned-pose success over broad continuation.

AI 总结：该研究针对热带气旋风速数据时空稀疏且无法近实时获取的问题，提出神经网络从静止卫星影像预测二维风场和强度指标，经 ablation 研究证实联合预测可提升最大风半径反演效果，尤其适用于快速增强事件。

链接：https://arxiv.org/abs/2611.08255

机构：Earth Systems Lab（地球系统实验室）； University of Valencia（瓦伦西亚大学）； University of Cambridge（剑桥大学）； Max Planck Institute for Marine Microbiology（马克斯·普朗克海洋微生物学研究所）； European Space Agency（欧洲空间局）； Imperial College London（伦敦帝国学院）

作者：Simon Donike, Pritthijit Nath, Cristina Radin, Arthur Avenas, Ruben Cartuyvels, Emiliano Diaz, William K. Jones, Anna Jungbluth

英文摘要：High-resolution tropical cyclone surface wind speeds derived from synthetic aperture radar (SAR) observations can support storm structure and intensity modelling, but remain spatially and temporally sparse and unavailable near-real-time. Geostationary satellites capture cloud tops continuously in space and time, but do not directly resolve surface winds. Here, we present simple neural networks to predict 2D wind fields and storm-scale intensity metrics from geostationary imagery. Through a series of ablation studies, we find that joint prediction of 2D fields and scalar metrics from the model latent space improves the retrieval of the (radius of) maximum wind, especially for rapid intensification events.

AI 总结：AutoCompress是一种硬件引导的LLM压缩框架，将延迟、吞吐量等硬件约束融入可微架构搜索，在多硬件平台及场景下较Minitron剪枝框架提升了准确率，还成功压缩了Nemotron 3.5 Super模型并保留了95%的基准分数。

链接：https://arxiv.org/abs/2611.08290

作者：Sheldon Liang, Ali Taghibakhshi, Saurav Muralidharan, Sharath Turuvekere Sreenivas, Ashwath Aithal, Pavlo Molchanov, Nima Tajbakhsh

英文摘要：LLM pruning must preserve quality while meeting hardware- and workload-specific constraints on latency, throughput, and memory. Existing pruning workflows often target parameter counts, requiring practitioners to manually select model sizes that meet latency, throughput, and GPU memory limits. However, parameter count alone does not predict these metrics, which depend on the model architecture, workload, hardware, and serving engine. We introduce AutoCompress, a hardware-guided compression framework that directly incorporates such constraints into differentiable architecture search. AutoCompress jointly optimizes global and per-layer structure to produce heterogeneous architectures, guided by serving-engine cost models validated against end-to-end latency, throughput, and memory measurements. We evaluate AutoCompress on long-context local coding, batched datacenter serving, and latency-sensitive in-game interaction. Compared to the state-of-the-art Minitron pruning framework, AutoCompress improves average accuracy across eight benchmarks by 6.57, 2.35, and 0.57 percentage points on RTX 5070, RTX 5080, and H100, respectively. AutoCompress also compresses Nemotron 3.5 Super from 120B to 74B parameters for single-B200 BF16 serving at 128K context, retaining 95% of its mean benchmark score.

AI 总结：该研究提出方向原子簇展开（dACE）模型，可从少量轨迹数据逐阶恢复集体运动的多体相互作用规律，其性能优于SocialNet，且学习到的规律具有可解释性。

链接：https://arxiv.org/abs/2611.08314

机构：School of Computer Science, University of Leeds（利兹大学计算机学院）

作者：Timon S. Gutleb

英文摘要：Models of collective motion usually reduce social interactions to pairwise forces, despite empirical evidence for many-body effects. We introduce the directional atomic cluster expansion (dACE), which adapts the atomic cluster expansion (ACE) of interatomic potentials to self-propelled agents. dACE provides a systematically improvable, exactly equivariant expansion of an agent's acceleration that splits uniquely into 1-, 2-, 3-body and higher contributions, along with a self term for the agent's own past motion that is fitted so that it does not absorb the interactions. On simulated collectives in two and three dimensions with known rules, dACE recovers the interaction law body order by body order from little data, while the neural model SocialNet, the current closest competitor, needs more data, has much larger errors on many-body or chiral rules and is restricted to 2D. On real data sets of tracked zebrafish and golden shiners, the laws learned by dACE agree better between disjoint halves of the data than SocialNet's. The learned laws are interpretable, and in the shiners the 3-body term opposes the pairwise response as neighbourhoods get dense. Finally, we show that next-step prediction accuracy mostly measures self-propulsion and noise rather than the interactions, and suggest more robust combinations of error measures for interaction learning.

AI 总结：StrataFlux框架通过协同设计精细化表示、选择与传递机制，在保留低比特量化内存与速度优势的同时恢复准确率，在多款模型与数据集上实现了性能与效率的显著提升。

链接：https://arxiv.org/abs/2611.08363

作者：Jake Hyun, Ziyu Zhao, Jae-sun Seo, Mohamed S. Abdelfattah

英文摘要： Quantization reduces the memory and bandwidth demands of large language model (LLM) inference but can degrade accuracy; offloading higher-precision data can instead stall generation. We introduce StrataFlux, a framework that challenges this tradeoff by recovering accuracy lost to low-bit weight and key-value (KV) cache quantization while retaining quantization's memory and speed advantages. StrataFlux co-designs refinement representation, selection, and delivery to make host RAM and solid-state drives useful sources of additional precision. Its selective precision streaming reduces transfer volume, overlaps delivery with computation, and reuses fetched precision across decoding steps so generation need not wait for every transfer. We evaluate 2-4-bit weights and 2-3-bit KV-cache on dedicated GPUs and unified-memory platforms. For 2-bit Llama-3.1-8B-Instruct weights, StrataFlux reduces WikiText-2 perplexity from 9.46 to 8.40 while maintaining 95.4% of the quantized base's throughput on RTX 5090. With 2-bit weights and KV, StrataFlux raises Qwen3-8B's AIME 2024/2025 accuracy from 11.0% to 47.1% at unchanged GPU memory use. At 128K context, Llama-3.1-8B-Instruct with 3-bit weights and 2-bit KV achieves comparable perplexity and speed to vLLM's 4-bit AWQ weights and 8-bit KV on RTX 5090, using approximately half the GPU memory. These results show that GPU-resident precision need not set the accuracy ceiling, nor offloading-tier latency dictate decoding speed.

AI 总结：该研究提出Muir模型，通过灵敏度分配低秩投影、匹配滤波读出门等改进线性注意力，在1.66亿至13.4亿参数下，降低困惑度、提升准确率与召回率，训练推理效率优于Transformer和GDN-2。

链接：https://arxiv.org/abs/2611.08376

机构：University of Texas at Austin（德克萨斯大学奥斯汀分校）

作者：Tushaar Gangavarapu, Gil Tabak, Clemens JS Schaefer

英文摘要：Linear attention compresses context into a fixed-size key-value state for linear-time training and constant-memory inference. Gated delta-rule models decay this state and erase its prediction for an incoming key before writing a new value; Gated DeltaNet-2 (GDN-2) generalizes this with channel-wise decay and decoupled, full-rank erase and write gates. However, we show that state updates are far less sensitive to erasure than to writes: key activation and unit normalization concentrate key energy onto a sparse subset of channels, whereas value writes scale all channels, so the erase gate can be parameterized at lower rank. Furthermore, slow-decaying state rows accumulate much larger norms than fast rows, dominating unweighted readout and forcing stronger write-time erasure to suppress interference. Guided by this analysis, we introduce Muir, which fuses all gates into a sensitivity-allocated low-rank projection, re-weights state rows with a matched-filter readout gate, and reinvests the saved parameters in a wider feed-forward network. Because standard chunkwise algorithms become numerically unstable at the large chunk sizes needed to saturate wide matrix units, we design a parallel update for stable, efficient training. Across 166M-1.34B parameters, Muir lowers perplexity by 1.8-2.0% over GDN-2 and 2.5-3.6% over Transformer, raises commonsense accuracy by up to 1.9 points, and improves extractive recall over GDN-2. At 1.34B, Muir outperforms both baselines on multi-hop and multi-needle recall, extrapolates 2,560x from its 4K training length to 10M-token streams without perplexity drift, and maintains near-100% passkey retrieval to 1M tokens with static readout. Muir achieves these gains while matching GDN-2 throughput, reaching its 100B-token validation loss 1.2x faster, and training 2.7x faster than Transformer at 32K.

AI 总结：本研究提出可解释的混合特征融合框架，结合LSTM自动编码器与类别加权XGBoost，在53750个NASA DASHlink飞行数据上实现着陆阶段罕见安全事件的平衡检测，平均宏召回率超0.90，为模型适配机场数据提供指导。

链接：https://arxiv.org/abs/2611.08386

作者：Taha Asghar, Bilal Mufti, Syed Qasim Zaheer, Luqman Ahmed

英文摘要：Landing-phase safety-event prediction requires the detection of rare abnormal approaches within strongly imbalanced multivariate flight data. Capturing coupled temporal behavior while distinguishing event-specific signatures across airport operating contexts remains a central modeling challenge. This study proposes an interpretable hybrid feature-fusion framework for multiclass landing-phase safety-event prediction. A long short-term memory (LSTM) autoencoder trained on nominal flight windows characterizes deviations from routine approach behavior. Its continuous reconstruction-error descriptor is fused with parameter-specific statistical flight-history features and supplied to a class-weighted XGBoost classifier to distinguish Normal, Speed High, Path High, and Flaps Late windows. The framework is evaluated using 53,750 NASA DASHlink approach windows from three arrival airports. Pooled and airport-specific learning strategies are compared alongside component-level benchmarks, repeated pipeline retraining, and feature-attribution analysis. The results demonstrate strong class-balanced detection, with mean macro recall above 0.90 across all airports under repeated partitioning and retraining. Component comparisons establish the importance of class weighting and show that continuous feature fusion outperforms reconstruction-error-only and hard-gate alternatives. Airport-level comparisons reveal benefits from local specialization at the two airports with larger datasets and from pooled training at the airport with the smallest dataset. Feature attributions associate predictions with operationally relevant vertical-path, configuration, speed, and approach-context variables. The framework provides an interpretable approach to rare-event detection and practical guidance for adapting flight-data monitoring models to airport-level data availability.

AI 总结：该研究系统调查分子表示的质量，发现先进分子嵌入模型在捕捉图结构等方面未优于指纹，经典指纹也存在非唯一表示的局限，为开发更稳健的小分子编码指明方向。

链接：https://arxiv.org/abs/2611.08420

机构：University of Oxford（牛津大学）

作者：Acer Blake, Fergus Imrie, Charlotte M. Deane

英文摘要：Molecular representations such as fingerprints and learned embeddings play a central role in cheminformatics and small-molecule machine learning. They serve as the primary input for a diverse array of tasks, including clustering, dataset splitting, and virtual screening, yet their evaluation is often limited to downstream model performance on narrow benchmark tasks. We carried out a systematic investigation of the quality of molecular representations, finding that none of the state-of-the-art molecular embedding models capture graph structure, 3D shape, or ADMET endpoint similarity better than fingerprints or physicochemical descriptors. However, classical fingerprints also revealed their own material limitations. For example, we show that depending on the representation, between 225M and $\sim$4B molecules have non-unique representations in the 10.8B Enamine REAL database, with implications for virtual screening. Our findings provide an insight into the current limitations of molecular representations, outlining future research directions for the development of more robust encodings of small molecules.

AI 总结：提出隐马尔可夫偏好模型，解决攻击者行为模式切换及行动参数时变问题，在CAGE 2基准上验证其性能优于循环基线。

链接：https://arxiv.org/abs/2611.08445

机构：SRI International（SRI国际）

作者：Angel Daruna, Michael Cogswell, Tixiao Shan, Yi Yao, Han-Pang Chiu, Ajay Divakaran

英文摘要：A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs. Two structures complicate this: the attacker's behavioral regime persists and can switch within an episode, and each action combines an operation with state-dependent arguments whose feasible values change over time. We capture both with a hidden Markov model whose latent state is the active regime and whose regime-conditioned action law is defined on a feasibility-masked tree of executable actions. The same likelihood forms the next-action distribution and updates the regime belief once the action is seen, and a partially observed prefix contributes its resolved local factors. Estimation separates into convex logistic fits that Baum-Welch reuses when regimes are unlabeled. We derive a reach-weighted nodewise KL expression for regime information, a sequence-wise log-loss bound for switching regime paths, and a uniform fallback guarantee. On the CAGE 2 cyber defense benchmark the filtered latent state is worth 0.8 nats per step, and after a controller switch the loss begins to fall in three steps against ten for a recurrent baseline.

AI 总结：本文针对整合信息理论的整合信息φ，在4、5单元随机网络中探究有限实验下机制量的可识别性、连续性与可证实性，发现行为无差异但内部可区分、部分访问有精度下限、φ对边界敏感，且可通过大量受控转移证实下界。

链接：https://arxiv.org/abs/2611.08469

机构：University of Waterloo（滑铁卢大学）； Perimeter Institute for Theoretical Physics（圆周理论物理研究所）； University of Debrecen（德布勒森大学）

作者：Linqian Wu, Diaoxuan He

英文摘要：Interpretability, AI auditing and assessments of machine consciousness increasingly rest on mechanism-level quantities: functionals of a system's causal transition law that describe how it computes, not what it outputs. Experiments see that law only through finite, partial and noisy interventions. We ask when such a quantity is identifiable from an experiment's access, continuous in the mechanism and certifiable from finite data, and answer exactly for a deliberately demanding case - the integrated information $\varphi$ of integrated information theory - in stochastic networks of four and five units with known ground truth. Behavior can be silent: a recurrent and a feed-forward network with identical input-output laws at their response delay carry 0.28 and 0 bits, yet internal recordings separate them within 4-53 trials. Partial access can be just as blind: in our examples, units left out of a one-step recording impose precision floors of 0.30-1.02 bits that no sample size lowers, and in all-but-self parity networks, recordings of any length see several hidden units only through the product of their fidelities - a symmetry that one intervention broke in exact checks. The measure has sharp edges: because $\varphi$ selects a partition, and the cause and effect it specifies, by one criterion and reports another, infinitesimal changes of mechanism that cross an edge change it by 25-100% at 99% of the edges we found in random networks, and about 1% of surrogate models that fit interventional data almost perfectly misreport it by more than 20%. One-sided claims survive: the same data certify a lower bound on both sides of an edge, and we prove $\varphi>1.004$ bits throughout a 64-dimensional region of mechanisms and certify it from about $10^7$ controlled transitions with false-certificate probability at most $\alpha$. Measurability is thus a design question with computable answers.

AI 总结：本文提出用户嵌入是可解释行为模式的稀疏叠加，用稀疏自编码器SAEs等可恢复模式，经合成数据及真实数据集验证，可实现零样本受众检索，且框架模型无关无需修改上游系统。

链接：https://arxiv.org/abs/2611.08487

机构：Amazon（亚马逊）

作者：Jonathan Kozaczuk, Zheng Dong, Noyan Evirgen, Keivan Majidi, Eddie K. H. Ng, Chris Porter, Qi Zhao, Jia Geng, Changmao Li, Brandon R. Feng

英文摘要：User embeddings are central components of modern recommendation, personalization, and behavioral modeling systems. Though these embeddings can compactly encode rich behavioral information, they are inherently dense and opaque, limiting interpretability. To address this gap, we argue that user embeddings can be understood as sparse superpositions of interpretable behavioral modes. This structure is directly predicted by generative models of user behavior for linear embeddings and extends approximately to deep-learning representations by the superposition hypothesis. Superimposed modes can naturally be recovered by sparse dictionary learning methods, such as sparse autoencoders (SAEs), and recovered modes can be automatically named and queried in natural language. We validate mode recovery against known ground truth on synthetic data, where SAEs recover modes better than classical sparse baselines, including on transformer-encoded embeddings. We then apply the method to classical and transformer-based user embeddings from two real-world datasets. We also show how the corresponding natural-language labels enable zero-shot audience retrieval and characterization over embeddings never trained on text. The framework is post-hoc and model-agnostic, requiring no modification to the upstream embedding system.