论文检索

General Machine Learning · Transfer, Multitask and Meta-learning

HuiYu Yi, Xu Zhiming, Dunwei Tu, Zhicheng Wang, Baile Xu, Furao Shen

The Nearest Class Mean (NCM) classifier is widely favored in Class-Incremental Learning (CIL) for its superior resistance to catastrophic forgetting compared to Fully Connected layers. While Neural Collapse (NC) theory supports NCM's optimality by assuming features collapse into single points, non-linear feature drift and insufficient training in CIL often prevent this ideal state. Consequently, classes manifest as complex manifolds rather than collapsed points, rendering the single-point NCM suboptimal. To address this, we propose Hierarchical-Cluster SOINN (HC-SOINN), a novel classifier that captures the topological structure of these manifolds via a ``local-to-global'' representation. Furthermore, we introduce Structure-Topology Alignment via Residuals (STAR) method, which employs a fine-grained pointwise trajectory tracking mechanism to actively deform the learned topology, allowing it to adapt precisely to complex non-linear feature drift. Theoretical analysis and Procrustes distance experiments validate our framework's resilience to manifold deformations. We integrated HC-SOINN into seven state-of-the-art methods by replacing their original classifiers, achieving consistent improvements that highlight the effectiveness and robustness of our approach. Code is available at \url{https://anonymous.4open.science/r/icml2026-9B60}.

Theory · Everything Else

Franz Nowak, Reda Boumasmoud, Ryan Cotterell

Recent progress in language model reasoning capabilities has revived a classic goal: characterizing which algorithms such systems can implement, or equivalently, which formal languages they can recognize. A growing research program studies RNNs, LSTMs, SSMs, and related architectures via reductions to finite-state automata and regular languages. Yet many results are not well-posed: they rely on implicit or incompatible semantic assumptions, invoking associativity and real-arithmetic techniques while also assuming floating-point arithmetic, which is finite and non-associative. Moreover, many proofs are highly architecture-specific and hard to transfer across closely related models. We address these issues with a unifying algebraic framework for a broad class of RNN language models, formally translating them to wreath products of transformation semigroups. By separating universal algebraic structure from contingent choices such as numerical semantics and wiring, the framework yields a disciplined workflow for rigorous expressivity analysis under realistic assumptions. We illustrate its value by rederiving and correcting representative expressivity claims from the literature under explicit deterministic finite-precision semantics.

Deep Learning · Large Language Models

Yan Wang, Chang Si, Kaiming Yang, Zhipeng Zhang, Weijian Liu, Man Yuan, Mingzhen Li, Yong Li, Weile Jia

Multi-token prediction (MTP) architecture is widely adopted in LLMs. MTP blocks can be appended to the tail of model to predict additional tokens. However, when training with pipeline parallel, MTP leads to more pipeline bubbles and deteriorates the pipeline efficiency. Based on in-depth analysis of MTP architectures and loss functions, we have identified the parallel nature of the MTP blocks, and leverage it for superior pipeline scheduling. We propose AdaHC, an adaptive pipeline scheduling framework for accelerating LLMs training with MTP block(s). AdaHC splits the output heads into chunks and reassembles the chunks to generate balanced pipeline stages, and performs adaptive activation forwarding to preserve the numerical equivalence. Experimental results show that AdaHC improves the training throughput of SOTA LLMs with diverse MTP configurations by 1.35$\times$ on average. This work paves a new direction for practical pipeline training.

Deep Learning · Large Language Models

Pingjie Wang, Hongcheng Liu, Yusheng Liao, Ziqing Fan, Yaxin Du, shuo tang, Yanfeng Wang, Yu Wang

Adapting large language models (LLMs) to low-resource domains remains challenging due to the scarcity of domain-specific data. While in-domain data is limited, there exists a vast amount of general-domain data that shares similar question–answer formats and reasoning patterns with domain tasks. This observation raises an important question: can useful general-domain data be mined to improve low-resource domain adaptation? Our initial findings show that general-domain chain-of-thought data contains useful auxiliary signals for domain adaptation, even without careful selection. This observation motivates a new paradigm for domain adaptation beyond exclusive reliance on domain-specific data. To systematically identify the most beneficial general-domain samples, we propose NTK-Selector, motivated by the Neural Tangent Kernel’s ability to capture alignment in training dynamics. Since directly applying NTK to pretrained LLMs is impractical, we introduce a Jacobian-free NTK approximation and empirically demonstrate stable NTK-like behavior during fine-tuning. Extensive experiments across medical, financial, legal, and psychological domains demonstrate that NTK-Selector consistently outperforms domain-only fine-tuning and existing data selection baselines. In particular, NTK-Selector achieves gains of +8.7 and +5.1 points on Llama3-8B-Instruct and Qwen3-8B, respectively, compared to only +0.8 and +0.9 points from domain-only fine-tuning.

Deep Learning · Large Language Models

Yuhan Liu, Yixiong Zou, Yuhua Li, Ruixuan Li

Referring Expression Segmentation (RES) aims to generate pixel-wise segmentation masks from complex and implicit textual queries. While recent advances in Multimodal Large Language Models (MLLMs) have substantially boosted RES performance, their prohibitive computational overhead remains a critical bottleneck, which, however, is rarely explored. To fill this gap, we first evaluate typical token compression methods on this task and observe a surprising performance degradation. In this paper, we aim to understand this phenomenon for a solution. By extensive experiments, we find that token compression for RES requires preserving the original position embeddings and local neighboring spatial structures, indicating that visual token position information is far more critical than in other tasks. Building on this insight, we ask: Can we design the token compression method purely based on the position information? Therefore, we propose PAYN, a plug-and-play, training-free token compression method that relies solely on position information. PAYN retains tokens that are adequately distributed in every local neighboring region while strictly preserving original positional indices, thereby maintaining spatial relational consistency. Experiments on multiple RES benchmarks demonstrate that our method outperforms existing token compression methods, verifying that position is indeed all you need for token compression in the MLLM-based RES task. Codes will be released.

Social Aspects · Safety

Joonhyuk Baek, Wonjune Seo, Jae-yun Kim, Saerom Park, Hoki Kim

Despite the proliferation of proactive defenses against deepfakes, the lack of a unified evaluation protocol precludes fair comparison and masks critical vulnerabilities. To bridge this gap, we present the first comprehensive benchmark that systematically assesses disruption, robustness, and transferability encompassing pixel, perceptual, and identity metrics. Our extensive analysis reveals that fidelity and identity metrics capture orthogonal performance axes, often leading to conflicting interpretations when relied upon individually. Furthermore, we identify a fundamental trade-off where peak white-box performance signals overfitting, and we introduce a calibrated evaluation to correct generator-induced identity bias. By exposing these blind spots, we establish a rigorous standard to guide the development of genuinely generalizable protections.

Applications · Chemistry, Physics, and Earth Sciences

Martin Jankowiak, Yerdos Ordabayev, Rudraksh Tuwani, Henry Ward, Hunter Nisonoff, James McFarland, Gevorg Grigoryan

Despite its importance to applications in protein design, predicting protein properties like binding affinity and thermostability from sparse experimental data remains a significant challenge. Accordingly, we introduce a class of sequence kernels that exploit evolutionary substitution matrices as well as local linearity and demonstrate that the resulting Gaussian processes provide data-efficient models of protein property landscapes, frequently outperforming alternatives that rely on foundation model embeddings. Furthermore--by learning what are in effect structure-aware substitution matrices--we show that our kernels can readily incorporate structural information from foundation models. We demonstrate that these structure-conditioned kernels are well suited to multi-task learning across multiple protein property landscapes and can decisively outperform local supervised learning methods.

Social Aspects · Accountability, Transparency, and Interpretability

Ziming Mao, Jia Xu, Wenxuan Pan, Mufan Xue, Yaochu Jin, Guoyuan Yang

Understanding the internal mechanisms of Deep Neural Networks remains a significant challenge, particularly in elucidating how generic visual concepts emerge within latent spaces. In this work, we propose SAEs-BrainMap, a novel framework that utilizes human brain activation patterns from the ventral visual pathway as objective probes to guide the identification of features decomposed by Sparse Autoencoders (SAEs). Our quantitative and qualitative empirical results demonstrate a robust representational alignment between sparse model features and biological Regions of Interest (ROIs), confirming the feasibility of utilizing brain signals to characterize model functionality. By leveraging this alignment, we trace the hierarchical trajectory of generic concepts cross layers and utilize the brain's hierarchical structure to visualize the model's global processing flow, providing novel insights into model interpretability.

Deep Learning · Generative Models and Autoencoders

Chengtao Lv, Yumeng Shi, Yushi Huang, Ruihao Gong, Shen Ren, Wenya Wang

Advanced autoregressive (AR) video generation models have improved visual fidelity and interactivity, but the quadratic complexity of attention remains a primary bottleneck for efficient deployment. While existing sparse attention solutions have shown promise on bidirectional models, we identify that applying these solutions to AR models leads to considerable performance degradation for two reasons: isolated consideration of chunk generation and insufficient utilization of past informative context. Motivated by these observations, we propose \textsc{Light Forcing}, the \textit{first} sparse attention solution tailored for AR video generation models. It incorporates a \textit{Chunk-Aware Growth} mechanism to quantitatively estimate the contribution of each chunk, which determines their sparsity allocation. This progressive sparsity increase strategy enables the current chunk to inherit prior knowledge in earlier chunks during generation. Additionally, we introduce a \textit{Hierarchical Sparse Attention} to capture informative historical and local context in a coarse-to-fine manner. Such two-level mask selection strategy (\ie, frame and block level) can adaptively handle diverse attention patterns. Extensive experiments demonstrate that our method outperforms existing sparse attention in quality (\eg, 84.5 on VBench) and efficiency (\eg, $1.2\sim1.3\times$ end-to-end speedup). Combined with FP8 quantization and LightVAE, \textsc{Light Forcing} further achieves a $2.3\times$ speedup and 19.7\,FPS on an RTX~5090 GPU.

Deep Learning · Large Language Models

Hoang Anh Duy Le, Sahil Joshi, Zeyu Yang, Zhaozhuo Xu, Anshumali Shrivastava

Self-attention dominates the computational and memory cost of long-context LLM inference across both prefill and decode phases. To address this challenge, we introduce **Sketch\&Walk** Attention, a training-free sparse attention method that determines sparsity with lightweight sketches and deterministic walk. Sketch\&Walk applies Hadamard sketching to get inexpensive approximations of attention scores, then aggregates these estimates across layers via a walk mechanism that captures attention influence beyond direct interactions between tokens. The accumulated walk scores are used to select top-$k$ attention blocks, enabling dynamic sparsity with a single training-free algorithm that applies uniformly to both the prefill and decode phases, together with custom sparse attention kernels. Across a wide range of models and tasks, Sketch\&Walk maintains near-lossless accuracy at 20\% attention density and can slightly outperform dense attention in some settings, while achieving up to $6\times$ inference speedup.

Optimization · Convex

Sharan Vaswani, Yifan Sun, Reza Babanezhad

Recent work has analyzed the convergence of first-order methods under non-uniform smoothness assumptions that better model the loss landscape in machine learning tasks. We generalize this assumption to objectives whose curvature is an affine function of the objective value. This property is satisfied by a broad class of problems, including logistic regression, generalized linear models with a logistic link function, softmax policy gradient in reinforcement learning, and a class of neural networks. Under this assumption and gradient domination conditions, we establish a general convergence rate for the steepest descent method, and deterministic, diagonal variants of RMSProp and Adam. Our results imply that for logistic regression on separable data and the softmax policy gradient objective, sign GD converges linearly and is provably faster than GD. Furthermore, we show that for a class of two-layer neural networks on separable data, RMSProp and Adam can converge at a linear rate with a constant step-size and momentum parameter. Finally, we present a lower bound demonstrating that, under our assumption, RMSProp and Adam are provably faster than AdaGrad, AMSGrad, gradient descent, and heavy-ball momentum.

Deep Learning · Large Language Models

Nuanqiao Shan, Kairong Han, Xinpeng Dong, Kun Kuang

Diffusion Language Models (DLMs) have emerged as a flexible alternative to autoregressive (AR) models. They can decode tokens in any order, but the generation quality critically depends on the decoding strategy. Existing approaches predominantly rely on local heuristics, such as confidence or entropy, which may fail to capture sequence-level dependencies and the semantics in the context. To solve this problem, we propose Latent-aware Unmasking Guidance Search (\LUGS{}), a novel decoding framework that leverages the model's internal hidden states to guide the unmasking process. By incorporating latent-aware scores to compensate for the limitations of local heuristics such as confidence or entropy, \LUGS{} improves the model's performance. Extensive experiments on various downstream tasks demonstrate that our approach consistently outperforms existing baselines on LLaDA-8B-instruct and LLaDA-1.5 models. In Science and Reason tasks, \LUGS{} improved performance by more than 1\% on both base models. And \LUGS{} obtains an average improvement of 3.5\% in code generation. Remarkably, \LUGS{} outperforms the beam search baseline by more than 5\% on average using LLaDA-8B-Instruct on code tasks. These results highlight the potential of latent-aware guidance for advancing controllable and high-quality generation.

Deep Learning · Large Language Models

Yuchen Yan, Liang Jiang, Jin Jiang, Shuaicheng Li, zujie wen, Zhiqiang Zhang, JUN ZHOU, Jian Shao, Yueting Zhuang, Yongliang Shen

Large reasoning models achieve strong performance by scaling inference-time chain-of-thought, but this paradigm suffers from quadratic cost, context length limits, and degraded reasoning due to lost-in-the-middle effects. Iterative reasoning mitigates these issues by periodically summarizing intermediate thoughts, yet existing methods rely on supervised learning or fixed heuristics and fail to optimize when to summarize, what to preserve, and how to resume reasoning. We propose InftyThink+, an end-to-end reinforcement learning framework that optimizes the entire iterative reasoning trajectory, building on model-controlled iteration boundaries and explicit summarization. InftyThink+ adopts a two-stage training scheme with supervised cold-start followed by trajectory-level reinforcement learning, enabling the model to learn strategic summarization and continuation decisions. Experiments on DeepSeek-R1-Distill-Qwen-1.5B show that InftyThink+ improves accuracy by 21\% on AIME24 and outperforms conventional long chain-of-thought reinforcement learning by a clear margin, while also generalizing better to out-of-distribution benchmarks. Moreover, InftyThink+ significantly reduces inference latency and accelerates reinforcement learning training, demonstrating improved reasoning efficiency alongside stronger performance.

General Machine Learning · Supervised Learning

Yuxin Tian, Mouxing Yang, Yuhao Zhou, Jian Wang, Qing Ye, Tongliang Liu, Gang Niu, Jiancheng Lv

In pursuit of data privacy, federated learning (FL) collaboratively trains a global model by aggregating local models learned from decentralized data. However, FL heavily depends on high-quality labels, which are often impractical in the real world, leading to the federated label-noise (F-LN) problem. Unlike traditional noisy labels, the F-LN problem is exacerbated by the inherent heterogeneity of FL, where clients experience varying levels and types of label errors. In this study, we observe that the global model of FL exhibits slow memorization of noisy labels, suggesting its ability to maintain reliable predictions and robust representations in FL. Based on this insight, we propose a novel method termed Global Reviser for Federated Learning with Noisy Labels (FedGR) to improve the robustness of FL against the F-LN problem. Specifically, FedGR first leverages the label-noise-robust characteristics of the global model to filter and refine the noisy labels on each client using the sieving-and-refining module. Then, it regularizes local model training with the assistance of the global model through the following two modules: the globally revised exponential moving average (EMA) distillation module and the global representation regularization module. Extensive experiments on three widely used F-LN benchmarks demonstrate the superior performance of FedGR, outperforming seven state-of-the-art baselines even in complicated label-noise and data heterogeneity. The code will be released upon acceptance.

Deep Learning · Large Language Models

Yuantian Shao, Peisong Wang, Yuanteng Chen, Chang Xu, Zhihui Wei, Jian Cheng

Large language models (LLMs) have achieved remarkable success, but their rapidly growing scale imposes prohibitive costs in memory, computation, and energy. Post-training quantization (PTQ) is a promising solution for efficient deployment, yet achieving accurate W4A4 quantization remains an open challenge. While most existing methods are designed for INT4 formats, the emergence of MXFP4—a new FP4 format with various hardware support (NVIDIA, AMD, Intel)—raises questions about the applicability of current techniques. In this work, we establish a comprehensive benchmark of PTQ methods under the MXFP4 format. Through systematic evaluation, we find that methods like GPTQ consistently deliver strong performance, whereas rotation-based approaches, which are almost used by all state-of-the-art approaches, suffer from severe incompatibility with MXFP4. We further provide the first in-depth analysis of this conflict, tracing its root to a fundamental mismatch between MXFP4’s PoT (power-of-two) block scaling and the redistribution of outlier energy via global rotation. Building on this insight, we propose a simple yet effective block rotation strategy that adapts rotation-based methods to MXFP4, leading to substantial accuracy improvements across diverse LLMs. Our findings not only offer clear guidance for practitioners but also set a foundation for advancing PTQ research under emerging low-precision formats.

Deep Learning · Graph Neural Networks

Xiang Wu, Rong-Hua Li, Xunkai Li, Kangfei Zhao, Hongchao Qin, Guoren Wang

Graph coarsening reduces the size of a graph while preserving certain properties. Most existing methods preserve either spectral or spatial characteristics. Recent research has shown that preserving topological features helps maintain the predictive performance of graph neural networks (GNNs) trained on the coarsened graph but suffers from exponential time complexity. To address these problems, we propose a Scalable Topology-Preserving Graph Coarsening (STPGC) by introducing the concepts of graph strong collapse and graph edge collapse extended from algebraic topology. STPGC comprises three new algorithms, *GStrongCollapse*, *GEdgeCollapse*, and *NeighborhoodConing* based on these two concepts, which eliminate dominated nodes and edges while rigorously preserving topological features. We further prove that STPGC preserves the GNN receptive field and develop approximate algorithms to accelerate GNN training. Experiments in node classification of GNNs demonstrate the efficiency and effectiveness of STPGC.

Deep Learning · Attention Mechanisms

Jerry Yao-Chieh Hu, Hude Liu, Hong-Yu Chen, Weimin Wu, Han Liu

We prove that with linear transformations, both (i) two-layer self-attention and (ii) one-layer self-attention followed by a softmax function are universal approximators for continuous sequence-to-sequence functions on compact domains. Our main technique is a new interpolation-based method for analyzing attention’s internal mechanism. This leads to our key insight: self-attention is able to approximate a generalized version of ReLU to arbitrary precision, and hence subsumes many known universal approximators. Building on these, we show that two-layer multi-head attention or even one-layer multi-head attention followed by a softmax function suffices as a sequence-to-sequence universal approximator. In contrast, prior works rely on feed-forward networks to establish universal approximation in Transformers. Furthermore, we extend our techniques to show that, (softmax-)attention-only layers are capable of approximating gradient descent in-context. We believe these techniques hold independent interest.

Deep Learning · Large Language Models

Xin Guo, Rongjunchen Zhang, Guilong Lu, Xuntao Guo, Jia Shuai, Zhi Yang, Liwen Zhang

Large language models are becoming increasingly significant in financial applications. Nevertheless, prevailing benchmarks are largely dependent on simulated or generic data, which leads to a significant gap between reported performance and actual efficacy in real-world scenarios. To tackle this challenge, we present BizFinBench.v2, the first integrated offline and online benchmark built upon authentic user query-response data from both Chinese and U.S. equity markets. It comprises 28,860 questions across eight offline and two online tasks. Experimental results show that GPT-5 achieves a mere 61.5% accuracy, still failing to meet the practical business requirement (84.8%). Among the evaluated commercial models, DeepSeek-R1 exhibits superior investment efficacy. Error analysis grounded in real financial practice reveals persistent limitations in existing models. By overcoming the constraints of prior benchmarks, BizFinBench.v2 provides a substantiated foundation for advancing LLM deployment in the financial sector.

Xinyou Wang, Liang Hong, Jiasheng Ye, Zaixiang Zheng, Shujian Huang, Quanquan Gu

Proteins are shaped by gradual evolution under biophysical and functional constraints. Protein language models learn rich evolutionary constraints from large-scale sequence data, and discrete diffusion–based protein language models (e.g., DPLMs) have emerged as a promising framework for both understanding and generation. However, existing DPLMs typically rely on masking-based absorbing diffusion, which conflicts with a basic biological intuition: proteins evolve through accumulated edits rather than emerging from masked tokens. As a result, these frameworks lack explicit pretraining objectives for substitution and insertion/deletion (indel) operations, which in turn limits both optimization-style post-editing and flexible guided generation. To address these limitations, we present DPLM-Evo, an evolutionary discrete diffusion framework that explicitly predicts substitution, insertion, and deletion operations during denoising. \method decouples a fixed-length latent alignment space from the variable-length observed sequence space, making indel-aware generation tractable and enabling adaptive scaffold growth throughout the process with negligible computational overhead. To further align substitutions with real evolutionary dynamics, we introduce a contextual evolutionary noising kernel that induces biologically informed, context-dependent mutation patterns. Across tasks, \method improves sequence understanding and achieves state-of-the-art performance on ProteinGym in the single-sequence setting, while also enabling variable-length simulated evolution, guided generation, and post-editing or optimization of existing proteins via explicit edit trajectories.

Applications · Time Series

Weijian Li, Hong-Yu Chen, Nabeel Rehemtulla, Ved Shah, Dongho Kim, Dennis Wu, Qinjie Lin, Adam Miller, Han Liu

Time series foundation models (TSFMs) are increasingly adopted as general-purpose time series learners. Although their training corpora are vast, they exclude peta-scale astronomical time series that exhibit unique challenges (e.g., irregular sampling, multiple variates, and heteroskedasticity) and exist in immense quantities. We introduce $\texttt{StarEmbed}$, the first public benchmark for stellar time series observations ("light curves") on three downstream tasks: unsupervised clustering, supervised classification, and out-of-distribution (OOD) source detection. $\texttt{StarEmbed}$ integrates a catalog of expert-vetted light curves, totaling $\sim$40,000 labeled examples across seven astrophysical classes. We evaluate the zero-shot capabilities of three families of TSFMs ($\texttt{Moirai}$, $\texttt{Chronos}$, and $\texttt{Time-MoE}$) and a domain-specific transformer ($\texttt{Astromer}$). Our results demonstrate that the $\texttt{Chronos}$ family, despite being pre-trained on regularly sampled data, outperforms domain-specific baselines and yields state-of-the-art performance in clustering and OOD source detection. While they do not yet strictly surpass hand-crafted features in classification, TSFM models like the Chronos models demonstrate excellent generalization performance, marking a promising step toward universal foundation models in astronomy.