论文检索

Deep Learning · Generative Models and Autoencoders

Yifu Luo, Haoyuan Sun, Xinhao Hu, Penghui Du, Keyu Fan, Bo Li, Sinan Du, Xu Wan, Zhiyu Chen, Bo Xia 等

Recent Progress in post-training flow matching for text-to-image (T2I) generation with Group Relative Policy Optimization (GRPO) has demonstrated strong potential. However, it is hindered by a critical limitation: inaccurate advantage attribution. In this work, we argue that aggregating consecutive timesteps into a coherent `chunk' and shifting the policy optimization paradigm from GRPO's step level to the chunk level can effectively mitigate the negative impact of this issue. Building on this insight, we propose Group Chunking Policy Optimization (GCPO), the first chunk-level reinforcement learning approach for post-training flow matching. Extensive experiments demonstrate that GCPO achieves superior performance on both standard T2I benchmarks and preference alignment, with up to $43\%$ additional gains over GRPO, highlighting the promise of chunk-level policy optimization.

Reinforcement Learning · Online

Jongchan Park, Seungjun Oh, Seungho Baek, Yusung Kim

Unsupervised Reinforcement Learning (URL) aims to pre-train scalable, skill-conditioned policies without extrinsic rewards, serving as a foundation for downstream control tasks. Despite recent progress, we argue that current off-policy URL methods are limited by two critical, overlooked bottlenecks: (1) non-stationarity skill semantic and (2) brittle generalization. To address these challenges, we propose GenDa (Generalizable Data-efficient Agent), a unified framework for robust unsupervised reinforcement learning. First, we introduce a skill relabeling mechanism to mitigate non-stationarity and significantly improve sample efficiency for pretraining. Second, we propose a Complementary Information Bottleneck(CIB), encouraging the learned skill policy to focus on ego-centric features and become robust to distribution shifts for downstream tasks. Through various experiments, we demonstrate that GenDa significantly enhances the scalability of URL with superior generality and sample efficiency. Our source codes are available in the supplementary materials.

Theory · Online Learning and Bandits

Sanghoon Yu, Min-hwan Oh

We study linear contextual bandits under rare parameter updates: the learner may incorporate reward feedback into its parameter estimate only at a small number of update times, while still observing contexts online and selecting actions sequentially. This viewpoint clarifies a practical distinction that is often blurred in the literature: many "strictly batched" methods additionally restrict within-interval context adaptivity, meaning that the action rule inside an interval cannot depend on the sequence of realized contexts/actions in that interval (beyond the current round's context). For linear contextual bandits, we propose two practical algorithms with only $O(\log\log T)$ parameter updates. Our first algorithm BLCE-G attains minimax-optimal regret (up to polylogarithmic factors in $T$) simultaneously in both the small-$K$ and large-$K$ regimes under a static schedule. Our second algorithm BLCE removes the near G-optimal design step---a dominant computational bottleneck in prior strictly batched static-grid methods---yet preserves minimax-optimal regret and achieves the lowest known runtime complexity among optimal algorithms. We further extend these rare-update and computational principles to generalized linear contextual bandits. Overall, our results yield statistically optimal algorithms under $O(\log\log T)$ parameter updates that are also computationally efficient in practice.

Social Aspects · Alignment

Yitong Sun, Yao Huang, Teng Li, Ranjie Duan, Yichi Zhang, Xingjun Ma, Hui Xue', Xingxing Wei

Mixture-of-Experts (MoE) architectures have emerged as a popular paradigm for scaling Large Language Models (LLMs), enabling greater capacity with reduced computational cost by dynamically routing inputs to the most relevant experts based on learned patterns. However, this also introduces a critical vulnerability: Safety Sparsity, where safety capabilities concentrate in few experts, making them susceptible to adversarial bypassing. Meanwhile, conventional alignment methods apply uniform adaptations across all parameters, ignoring their functional differences and inadvertently degrading general utility. To address these challenges, we propose MESA (MoE Safety Alignment), a targeted alignment framework for MoE-based LLMs that strategically decentralizes safety responsibilities to maximize coverage while explicitly minimizing interference with general capabilities. Specifically, based on Optimal Transport (OT) theory, MESA operates through two mechanisms: (1) Expert Capacity Reallocation, which uses a transport cost matrix to distribute safety duties to the most cost-effective experts; and (2) Dynamic Routing Refinement, which constrains the router to ensure precise activation of these decentralized modules. Extensive experiments demonstrate that MESA achieves robust defensive performance against varied harmful benchmarks while preserving general helpfulness.

Applications · Computer Vision

Zhanhe Zhang, Jiahua Li, Kun Wei, Xu Yang, Cheng Deng

Video inpainting aims to restore missing regions while preserving spatial and temporal coherence. Diffusion-based methods achieve strong per-frame reconstruction, but their sampling implicitly generates temporally coupled latent trajectories whose long-horizon stability is not explicitly modeled, leading to a trade-off between temporal consistency and structural detail. We revisit video inpainting from the perspective of temporal trajectory stability, viewing temporal inconsistency as instability along time-indexed denoising trajectories rather than an output-level error. Based on this view, we propose an inference-time trajectory stabilization framework that monitors motion-aligned deviation and triggers risk-aware correction only when instability accumulates. It combines sparsely sampled trajectory anchors as stability references with neighborhood-consistent propagation to regulate trajectory evolution while preserving local generative freedom. Implemented as a lightweight control layer in the sampling loop, it selectively contracts unstable trajectories toward motion-consistent manifolds instead of enforcing uniform temporal constraints. Experiments show consistent improvements in temporal coherence and structural fidelity.

Theory · Optimization

Taha EL BAKKALI EL KADI, Rayane Bouftini, Richard Zhang, Omar Saadi

We consider minimizing high-dimensional smooth nonconvex objectives using only noisy pairwise comparisons. Unlike classical zeroth-order methods limited by the ambient dimension $d$, we propose Noisy-Comparison Random Search (NCRS), a direct-search method that exploits random line search to adapt to the intrinsic dimension $k \le d$. We establish a novel nonconvex analysis for approximate stationarity: under a uniform-margin oracle with advantage $p$, NCRS attains $\epsilon$-stationarity with complexity $\mathcal{O}(k/(p^{2}\epsilon^{2}))$, explicitly replacing ambient dependence with the intrinsic dimension. Furthermore, we introduce a general tie-aware noise model where comparison quality degrades near ties; for this setting, we prove that a majority-vote variant of NCRS achieves $\epsilon$-stationarity with complexity $\mathcal{O}(k^{2}/\epsilon^{4})$.

Theory · Deep Learning

Kaifei Wang, Binghui Li, Han Zhong, Pinyan Lu, Liwei Wang

Muon updates matrix parameters via the matrix sign of the gradient and has shown strong empirical gains, yet its dynamics and scaling behavior remain unclear in theory. We study Muon in a linear associative memory model with softmax retrieval and a hierarchical frequency spectrum over query–answer pairs, with and without label noise. In this setting, we show that Gradient Descent (GD) learns frequency components at highly imbalanced rates, leading to slow convergence bottlenecked by low-frequency components. In contrast, the Muon optimizer mitigates this imbalance, leading to faster and more uniform progress. Specifically, in the noiseless case, Muon achieves an exponential speedup over GD; in the noisy case with a power-decay frequency spectrum, we derive Muon's optimization scaling law and demonstrate its superior scaling efficiency over GD. Furthermore, we show that Muon can be interpreted as an implicit matrix preconditioner arising from adaptive task alignment and block-symmetric gradient structure. In contrast, the preconditioner with coordinate-wise sign operator could match Muon under oracle access to unknown task representations, which is infeasible for SignGD in practice. Experiments on synthetic long-tail classification and LLaMA-style pre-training corroborate the theory.

Reinforcement Learning · Deep RL

Zihan Lin, Xiaohan Wang, Jie Cao, Jiajun Chai, Li Wang, Xiaodong Lu, Wei Lin, Ran He, Guojun Yin

Reinforcement Learning with Verifiable Rewards (RLVR) enhances reasoning of Large Language Models (LLMs) but usually exhibits limited generation diversity due to the over-incentivization of positive rewards. Although methods like Negative Sample Reinforcement (NSR) mitigate this issue by upweighting penalty from negative samples, they may suppress the semantic distributions shared between positive and negative responses. To boost reasoning ability without losing diversity, this paper proposes negative sample projection Residual Reinforcement Learning (ResRL) that decouples similar semantic distributions among positive and negative responses. We theoretically link Lazy Likelihood Displacement (LLD) to negative-positive head-gradient interference and derive a single-forward proxy that upper-bounds representation alignment to guide conservative advantage reweighting. ResRL then projects negative-token hidden representations onto an SVD-based low-rank positive subspace and uses projection residuals to modulate negative gradients, improving reasoning while preserving diversity and outperforming strong baselines on average across twelve benchmarks spanning Mathematics, Code, Agent Tasks, and Function Calling. Notably, ResRL surpasses NSR on mathematical reasoning by 9.4\% in Avg@16 and 7.0\% in Pass@128. Code is available at https://anonymous.4open.science/r/ResRL.

Zhenyu Wu, Yao Huang, Shouwei Ruan, Xingxing Wei

Text-to-image diffusion models have achieved remarkable success in generating high-quality images, yet existing safety mechanisms exhibit critical cross-seed instability where defense performance varies significantly under different random seed conditions. This instability stems from the fact that a single malicious prompt generates diverse harmful variants across different noise initializations, forming complex distributional clusters that current methods cannot adequately address. We investigate extending Noise Contrastive Alignment (NCA) to diffusion models due to its native capability of handling multiple negative samples through probabilistic weighting, but our theoretical analysis reveals two fundamental flaws in direct extension: gradient reversal caused by positive regularization terms that paradoxically penalize safe content generation, and uniform suppression of harmful samples that ignores severity variations. To tackle these issues, we propose Noise Contrastive Diffusion (NCD), which incorporates targeted algorithmic modifications including elimination of problematic regularization and introduction of pairwise regularization mechanisms that establish individualized preference relationships between safe and harmful variants. Extensive experiments further demonstrate that NCD achieves superior cross-seed stability, reducing attack success rates (ASRs) from 11.1% to 6.2% compared to SOTA methods at the seed level while maintaining exceptional generation quality, exhibiting robust resistance against sophisticated jailbreak prompts and strong generalizability across different T2I architectures. WARNING: This paper may contain examples of harmful texts and images.

Applications · Computer Vision

Sheng Jiang, Lin Zhu, Runrui Li, Mei Wang, Qiannan Zhu, Yaoyao Zhong, Hua Huang

Handwritten mathematical expression recognition (HMER) remains challenging in real-world educational scenarios, even with recent advances in large vision-language models. While these models often achieve high accuracy in local symbol transcription, their reliability in capturing two-dimensional mathematical structure under realistic handwritten conditions is still poorly understood. We introduce a real-world handwritten benchmark covering 13 categories of structurally complex expressions with authentic writing artifacts. Evaluations on large models reveal a clear performance degradation as structural complexity increases, even when symbol-level accuracy is high. Most failures arise from structural mis-parsing and context-dependent symbol role confusion rather than pure visual perception errors. To mitigate this issue, we propose a training-free, schema-anchored structure-aware inference framework that decomposes recognition into schema identification, schema-constrained transcription, and context-driven disambiguation. Our method improves the ExpRate from 11.63\% to 24.52\% on Qwen-8B and generalizes well across multiple large models. Our benchmark provides a realistic evaluation for large models on handwritten mathematics, and our framework offers an effective and interpretable solution to structure-related failures in real-world HMER.

Deep Learning · Large Language Models

Chungpa Lee, Jy-yong Sohn, Kangwook Lee

Transformer-based large language models exhibit in-context learning, enabling adaptation to downstream tasks via few-shot prompting with demonstrations. In practice, such models are often fine-tuned to improve zero-shot performance on downstream tasks, allowing them to solve tasks without examples and thereby reducing inference costs. However, fine-tuning can degrade in-context learning, limiting the performance of fine-tuned models on tasks not seen during fine-tuning. Using linear attention models, we provide a theoretical analysis that characterizes how fine-tuning objectives modify attention parameters and identifies conditions under which this leads to degraded few-shot performance. We show that fine-tuning all attention parameters can harm in-context learning, whereas restricting updates to the value matrix improves zero-shot performance while preserving in-context learning. We further show that incorporating an auxiliary few-shot loss enhances in-context learning primarily on the target task, at the expense of degraded in-context learning ability on tasks not seen during fine-tuning. We empirically validate our theoretical results.

Wanting Liang, Haoang Chi, Zhiheng Zhang

Estimating Individual Treatment Effects (ITE) in multi-treatment scenarios faces two critical challenges: the Hyperparameter Selection Dilemma for balancing weights and the Curse of Dimensionality in computational scalability. This paper derives a novel multi-treatment generalization bound and proposes a theoretical estimator for the optimal balancing weight $\alpha$, eliminating expensive heuristic tuning. We investigate three balancing strategies: Pairwise, One-vs-All (OVA), and Treatment Aggregation. While OVA achieves superior precision in low-dimensional settings, our proposed Treatment Aggregation ensures both accuracy and $\mathcal{O}(1)$ scalability as the treatment space expands. Furthermore, we extend our framework to a generative architecture, Multi-Treatment CausalEGM, which preserves the Wasserstein geodesic structure of the treatment manifold. Experiments on semi-synthetic and image datasets demonstrate that our approach significantly outperforms traditional models in estimation accuracy and efficiency, particularly in large-scale intervention scenarios.

Deep Learning · Self-Supervised Learning

Woojun Jung, Susik Yoon

Tabular data within a domain often exhibit heterogeneous schemas yet shared semantics, posing a key challenge: determining what should remain invariant across tables and what should preserve instance-level distinctions. Existing token- or row-centric encoders conflate these roles, leading to schema sensitivity or weakened discriminability. We introduce the segment, a header–value pair, as an atomic unit that captures both functional roles and semantic content. Using value entropy, we treat low-entropy segments as domain anchors and high-entropy segments as entity-specific signals. We realize this design through Masked Segment Modeling and Entropy-driven Segment Alignment, which jointly enforce structured header–value coupling and selective semantic alignment. Experiments on in-domain heterogeneous tables demonstrate improved performance on discriminative and generative tasks, yielding stable and interpretable representations.

Applications · Robotics

Dongsheng Wang, Dawei Su, Hui Huang

Recently, zero-shot 3D scene understanding via 2D Vision-Language Models (VLMs) has gained increasing research interest due to their promising spatial reasoning capabilities. Typically, multiple 2D views are sampled from a 3D point cloud and fed into pre-trained VLMs to answer a given question. This paradigm highlights the critical role of input context quality and raises the challenge of retaining as many task-relevant 3D details as possible under a limited input budget. We propose \texttt{KeyVT}, a hierarchical approach for input context collection at both the view and token levels. Specifically, we combine pixel features with camera parameters and assess view importance based on both semantic content and geometric position, resulting in spatially consistent and task-relevant views. Furthermore, we address redundancy among patches across selected views by identifying representative tokens under the optimal transport (OT) framework, where view tokens and key tokens are formulated as two discrete distributions in the embedding space. These key tokens are expected to cover all view features by minimizing the OT distance. We evaluate our framework on three widely used benchmarks, demonstrating significant improvements over existing tuning-free methods and performance comparable to training-based approaches.

Deep Learning · Everything Else

Longhua Li, Lei Qi, Qi Tian, Xin Geng

While orthogonal subspace methods try to mitigate task interference in Continual Learning (CL), they often suffer from energy diffusion across the basis, hindering knowledge compaction and exhausting capacity for future tasks. We observe that output feature drift induced by parameter updates is inherently low-rank, and theoretically prove that preserving parameters along the principal directions of this drift minimizes the output reconstruction error. Motivated by this, we propose **E**nergy-Concentrated and **E**nergy-Ordered **Lo**w-**R**ank **A**daptation (E$^2$-LoRA). By explicitly ordering and concentrating knowledge into leading ranks, E$^2$-LoRA frees capacity for subsequent tasks. Furthermore, we design a dynamic rank allocation strategy to balance stability and plasticity by jointly optimizing energy retention and model plasticity. Extensive experiments across multiple benchmarks demonstrate that E$^2$-LoRA achieves state-of-the-art performance.

General Machine Learning · Scalable Algorithms

Haolin Yu, Guojun Zhang, Hongliang Li, Pascal Poupart

Federated representation learning (FRL) aims to learn personalized federated models with effective feature extraction from local data. FRL algorithms that share the majority of the model parameters face significant challenges with huge communication overhead. This overhead stems from the millions of neural network parameters and slow aggregation progress of the averaging heuristic. To reduce the overhead, we propose FedLog, which shares sufficient data summaries instead of raw model parameters. The data summaries encode minimal sufficient statistics of an exponential family, and Bayesian inference is utilized for global aggregation. FedLog helps reduce message sizes and communication frequency. We prove that the shared messages are minimal sufficient statistics and theoretically analyze the convergence rate of FedLog. To further ensure formal privacy guarantees, we extend FedLog with the differential privacy framework. Empirical results demonstrate high learning accuracy with low communication overhead of our method.

Applications · Neuroscience, Cognitive Science

Yiwen Gu, Junchuan Gu, Haibin Shen, Kejie Huang

Spiking Neural Networks (SNNs) emulate the spiking behavior of biological neurons and are promising for energy-efficient neuromorphic computing. A widely used strategy to train SNNs is to convert pretrained Artificial Neural Networks (ANNs), where the accuracy and efficiency are determined by the spike encoding scheme. Traditional methods based on spike count or timing severely underutilize the available encoding space, leading to large accuracy degradation under low-timestep constraints. More expressive alternatives involve complex dynamics, which hinder scalability and practical deployment. To address these challenges, we propose Temporal Weighted Encoding (TWE). Through a simple recursive integration, spikes are implicitly assigned exponentially decaying weights, drawing an analogy to a temporal bit sequence. We systematically analyze the temporal mismatch caused by this weight pattern and propose temporal relaxation and threshold relaxation to resolve this issue, enabling fast and accurate activation encoding. Extensive experiments demonstrate that TWE achieves negligible conversion loss with significantly fewer timesteps, offering a scalable and efficient solution for SNN deployment.

Si-Yang Liu, Han-Jia Ye

Tabular foundation models, exemplified by TabPFN, perform prediction via in-context learning, inferring test labels directly from labeled training examples. They have demonstrated competitive performance, particularly on small-to-medium datasets. However, recent tabular foundation models often improve accuracy with increasingly complex architectures, incurring higher inference cost and limiting practical deployment. In this work, we revisit the original TabPFN design and show that a lightweight row-wise attention–only backbone can remain highly competitive with two simple enhancements: a gated attention stabilization mechanism and a small set of learnable register tokens that provide global context and improve pretraining quality. The resulting model, SwiftPFN, supports both classification and regression, and is competitive with stronger tabular foundation models (e.g., TabPFN v2 and TabICL) while being more efficient at inference. For latency-sensitive serving, we further introduce an adaptive layer-wise early-exit mechanism that dynamically adjusts inference depth per sample. Experiments show that many samples can be reliably predicted using shallow layers, reducing average computation with negligible performance degradation. Overall, SwiftPFN enables efficient and anytime tabular in-context learning for practical deployments.

Hugo Tabanelli, Yatin Dandi, Luca Pesce, FLORENT KRZAKALA

Why depth yields a genuine computational advantage over shallow methods remains a central open question in learning theory. We study this question in a controlled high-dimensional Gaussian setting, focusing on compositional target functions. We analyze their learnability using an explicit three-layer fitting model trained via layer-wise spectral estimators. Although the target is globally a high-degree polynomial, its compositional structure allows learning to proceed in stages: an intermediate representation reveals structure that is inaccessible at the input level. This reduces learning to simpler spectral estimation problems, well studied in the context of multi-index models, whereas any shallow estimator must resolve all components simultaneously. Our analysis relies on Gaussian universality, leading to sharp separations in sample complexity between two and three-layer learning strategies.

Applications · Chemistry, Physics, and Earth Sciences

ZHIRAN HOU, TINGHUAI MA, Huan Rong, Li Jia, Anouar Imel, Heng Zhang, Ming Li

Molecular property prediction from 3D structures is fundamentally constrained by the scarcity of labeled data. To address this challenge, researchers have adapted various self-supervised pre-training methods from computer vision and natural language processing; however, these approaches often neglect the fundamental physical principles unique to molecular systems. When grounded in physical principles, denoising pre-training can be formally shown to be equivalent to learning molecular force fields.However, existing methods uniformly apply a uniform noise scheme across all molecules, which introduces systematic bias in molecular distribution modeling. To overcome this limitation, we propose MOES-Pred, a novel denoising pre-training framework that employs an energy sentinel mechanism to dynamically adjust molecule-specific noise perturbations. By incorporating chemical prior knowledge, we design molecule-specific noising strategies that expand conformational sampling coverage and improve the fidelity of molecular distribution modeling. Extensive experiments demonstrate that MOES-Pred consistently surpasses existing methods, achieving state-of-the-art performance on both force prediction tasks and downstream quantum chemical property predictions.