论文检索

General Machine Learning · Representation Learning

Daiki Yoshikawa, Takashi Matsubara

Vision-language models have achieved remarkable success in multi-modal representation learning from large-scale pairs of visual scenes and linguistic descriptions. However, they still struggle to simultaneously express two distinct types of semantic structures: the hierarchy within a concept family (e.g., *dog* $\preceq$ *mammal* $\preceq$ *animal*) and the compositionality across different concept families (e.g., "a dog in a car" $\preceq$ *dog*, *car*). Recent works have addressed this challenge by employing hyperbolic space, which efficiently captures tree-like hierarchy, yet its suitability for representing compositionality remains unclear. To resolve this dilemma, we propose *PHyCLIP*, which employs an $\ell_1$-*P*roduct metric on a Cartesian product of *Hy*perbolic factors. With our design, intra-family hierarchies emerge within individual hyperbolic factors, and cross-family composition is captured by the $\ell_1$-product metric, analogous to a Boolean algebra. Experiments on zero-shot classification, retrieval, hierarchical classification, and compositional understanding tasks demonstrate that PHyCLIP outperforms existing single-space approaches and offers more interpretable structures in the embedding space.

General Machine Learning · Representation Learning

Zijing Zhang, Ziyang Chen, Mingxiao Li, Zhaopeng Tu, Xiaolong Li

The development of autonomous agents for complex, long-horizon tasks is a central goal in AI. However, dominant training paradigms face a critical limitation: reinforcement learning (RL) methods that optimize solely for final task success often reinforce flawed or inefficient reasoning paths, a problem we term inefficient exploration. This leads to agents that are brittle and fail to generalize, as they learn to find solutions without learning how to reason coherently. To address this, we introduce RLVMR, a novel frame-work that integrates dense, process-level supervision into end-to-end RL by rewarding verifiable, meta-reasoning behaviors. RLVMR equips an agent to explicitly tag its cognitive steps—such as planning, exploration, and reflection—and provides program-matic, rule-based rewards for actions that contribute to effective problem-solving. These process-centric rewards are combined with the final outcome signal and optimized using a critic-free policy gradient method. On the challenging ALFWorld and ScienceWorld benchmarks, RLVMR achieves new state-of-the-art results, with our 7B model reaching an 83.6% success rate on the most difficult unseen task split. Our analysis confirms these gains stem from improved reasoning quality, including significant reductions in redundant actions and enhanced error recovery, leading to more robust, efficient, and interpretable agents.

Applications · Robotics

Pietro Mazzaglia, Cansu Sancaktar, Markus Peschl, Daniel Dijkman

Using Large Language Models to produce intermediate thoughts, a.k.a. Chain-of-thought (CoT), before providing an answer has been a successful recipe for solving complex language tasks. In robotics, similar embodied CoT strategies, generating thoughts before actions, have also been shown to lead to improved performance when using Vision-Language-Action models (VLAs). As these techniques increase the length of the model's generated outputs to include the thoughts, the inference time is negatively affected. Delaying an agent's actions in real-world executions, as in robotic manipulation settings, strongly affects the usability of a method, as tasks require long sequences of actions. However, is the generation of long chains-of-thought a strong prerequisite for achieving performance improvements? In this work, we explore the idea of Hybrid Training (HyT), a framework that enables VLAs to learn from thoughts and benefit from the associated performance gains, while enabling the possibility to leave out CoT generation during inference. Furthermore, by learning to conditionally predict a diverse set of outputs, HyT supports flexibility at inference time, enabling the model to either predict actions directly, generate thoughts or follow instructions. We evaluate the proposed method in a series of simulated benchmarks and real-world experiments.

Computer Vision · Image and Video Generation

Nupur Kumari, Sheng-Yu Wang, Cherry Zhao, Yotam Nitzan, Yuheng Li, Krishna Kumar Singh, Richard Zhang, Eli Shechtman, Jun-Yan Zhu, Xun Huang

Recent image editing models have achieved impressive results while following natural language editing instructions, but they rely on supervised fine-tuning with large datasets of input-target pairs. This is a critical bottleneck, as such naturally occurring pairs are hard to curate at scale. Current workarounds use synthetic training pairs that leverage the zero-shot capabilities of existing models. However, this can propagate and magnify the artifacts of the pretrained model into the final trained model. In this work, we present a new training paradigm that eliminates the need for paired data entirely. Our approach directly optimizes a few-step diffusion model by unrolling it during training and leveraging feedback from vision-language models (VLMs). For each input and editing instruction, the VLM evaluates if an edit follows the instruction and preserves unchanged content, providing direct gradients for end-to-end optimization. To ensure visual fidelity, we incorporate distribution matching loss (DMD), which constrains generated images to remain within the image manifold learned by pretrained models. We evaluate our method on standard benchmarks and include an extensive ablation study. Without any paired data, our method performs on par with various image editing diffusion models trained on extensive supervised paired data, under the few-step setting. Given the same VLM as the reward model, we also outperform RL-based techniques like Flow-GRPO.

Jaskirat Singh, Xingjian Leng, Zongze Wu, Liang Zheng, Richard Zhang, Eli Shechtman, Saining Xie

Representation alignment helps generation by distilling representations from a pretrained vision encoder to intermediate diffusion features. We investigate a fundamental question - `what aspect of the target representation matters for generation, its global information (measured by Imagenet1K accuracy) or its spatial structure (pairwise cosine similarity between patch tokens)''? Prevalent wisdom holds that stronger global performance leads to better generation as a target representation. To study this, we first perform a large-scale empirical analysis across 27 different vision encoders and different model scales. The results are surprising - spatial structure, rather than global performance drives the generation performance of a target representation. To further study this, we introduce two straightforward modifications, which specifically accentuate the transfer of spatial information. We replace the standard MLP projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation. Surprisingly, our simple method (implemented in <4 lines of code), termed iREPA, consistently improves convergence speed of REPA, across a diverse set of vision encoders, model sizes, and training variants (such as REPA, REPA-E, meanflow, JiT etc). Our work motivates revisiting the fundamental working mechanism of representational alignment and how it can be leveraged for improved training of generative models.

Optimization · Everything Else

Shuche Wang, Fengzhuo Zhang, Jiaxiang Li, Cunxiao Du, Chao Du, Tianyu Pang, Zhuoran Yang, Mingyi Hong, Vincent Tan

The Muon optimizer is consistently faster than Adam in training Large Language Models (LLMs), yet the mechanism underlying its success remains unclear. This paper demystifies this mechanism through the lens of associative memory. By ablating the transformer components optimized by Muon, we reveal that the associative memory parameters of LLMs, namely the Value and Output (VO) attention weights and Feed-Forward Networks (FFNs), are the primary contributors to Muon’s superiority. Motivated by this associative memory view, we then explain Muon’s superiority on real-world corpora, which are intrinsically heavy-tailed: a few 'head' classes are extremely frequent, while a vast number of 'tail' classes are individually rare. The superiority is explained through two key properties: (i) its update rule consistently yields a more isotropic singular spectrum than Adam; and as a result, (ii) on heavy-tailed data, it optimizes tail classes more effectively than Adam. Beyond empirical evidence, we theoretically confirm these findings by analyzing a one-layer associative memory model under class-imbalanced data. We prove that Muon consistently achieves balanced learning across classes regardless of feature embeddings, whereas Adam can induce large disparities in learning errors depending on embedding properties. In summary, our empirical observations and theoretical analyses reveal Muon’s core advantage: its update rule aligns with the outer-product structure of linear associative memories, enabling more balanced and effective learning of tail classes in heavy-tailed distributions than Adam.

Zhongjian Qiao, Rui Yang, Jiafei Lyu, Xiu Li, Zhongxiang Dai, Zhuoran Yang, Siyang Gao, Shuang Qiu

Single-domain offline reinforcement learning (RL) often suffers from limited data coverage, while cross-domain offline RL handles this issue by leveraging additional data from other domains with dynamics shifts. However, existing studies primarily focus on train-time robustness (handling dynamics shifts from training data), neglecting the test-time robustness against dynamics perturbations when deployed in practical scenarios. In this paper, we investigate dual (both train-time and test-time) robustness against dynamics shifts in cross-domain offline RL. We first empirically show that the policy trained with cross-domain offline RL exhibits fragility under dynamics perturbations during evaluation, particularly when target domain data is limited. To address this, we introduce a novel robust cross-domain Bellman (RCB) operator, which enhances test-time robustness against dynamics perturbations while staying conservative to the out-of-distribution dynamics transitions, thus guaranteeing the train-time robustness. To further counteract potential value overestimation or underestimation caused by the RCB operator, we introduce two techniques, the dynamic value penalty and the Huber loss, into our framework, resulting in the practical Dual-RObust Cross-domain Offline RL (DROCO) algorithm. Extensive empirical results across various dynamics shift scenarios show that DROCO outperforms strong baselines and exhibits enhanced robustness to dynamics perturbations.

Deep Learning · Generative Models and Autoencoders

Jiebin Zhang, Zhenghan Yu, Liang Wang, Nan Yang, Eugene Yu, Zheng Li, Yifan Song, Dawei Zhu, Xingxing Zhang, Furu Wei 等

Speculative decoding accelerates large language model (LLM) inference by using a small draft model to generate candidate tokens for a larger target model to verify. The efficacy of this technique hinges on the trade-off between the time spent on drafting candidates and verifying them. However, current state-of-the-art methods rely on a static time allocation, while recent dynamic approaches optimize for proxy metrics like acceptance length, often neglecting the true time cost and treating the drafting and verification phases in isolation. To address these limitations, we introduce Learning to Draft (LTD), a novel method that directly optimizes for throughput of each draft-and-verify cycle. We formulate the problem as a reinforcement learning environment and train two co-adaptive policies to dynamically coordinate the draft and verification phases. This encourages the policies to adapt to each other and explicitly maximize decoding efficiency. We conducted extensive evaluations on five diverse LLMs and four distinct tasks. Our results show that LTD achieves speedup ratios ranging from 2.24x to 4.32x, outperforming the state-of-the-art method Eagle3 up to 36.4\%.

Computer Vision · Everything Else

Zhenghao Peng, Yuxin Liu, Bolei Zhou

Realistic and interactive traffic simulation is essential for training and evaluating autonomous driving systems. However, most existing data-driven simulation methods rely on static initialization or log-replay data, limiting their ability to model dynamic, long-horizon scenarios with evolving agent populations. We propose SceneStreamer, a unified autoregressive framework for continuous scenario generation that represents the entire scene as a sequence of tokens, including traffic light signals, agent states, and motion vectors, and generates them step by step with a transformer model. This design enables SceneStreamer to continuously introduce and retire agents over an unbounded horizon, supporting realistic long-duration simulation. Experiments demonstrate that SceneStreamer produces realistic, diverse, and adaptive traffic behaviors. Furthermore, reinforcement learning policies trained in SceneStreamer-generated scenarios achieve superior robustness and generalization, validating its utility as a high-fidelity simulation environment for autonomous driving. More information is available at https://vail-ucla.github.io/scenestreamer/ .

Siyu Chen, Heejune Sheen, Xuyuan Xiong, Tianhao Wang, Zhuoran Yang

We study the challenge of achieving theoretically grounded feature recovery using Sparse Autoencoders (SAEs) for the interpretation of Large Language Models. Existing SAE training algorithms often lack rigorous mathematical guarantees and suffer from practical limitations such as hyperparameter sensitivity and instability. We rethink this problem from the perspective of neuron activation frequencies, and through controlled experiments, we identify a striking phenomenon we term neuron resonance: neurons reliably learn monosemantic features when their activation frequency matches the feature's occurrence frequency in the data. Building on this finding, we introduce a new SAE training algorithm based on ``bias adaptation'', a technique that adaptively adjusts neural network bias parameters to ensure appropriate activation sparsity. We theoretically prove that this algorithm correctly recovers all monosemantic features when input data is sampled from our proposed statistical model. Furthermore, we develop an improved empirical variant, Group Bias Adaptation (GBA), and demonstrate its superior performance against benchmark methods when applied to LLMs with up to 2 billion parameters. This work represents a foundational step in demystifying SAE training by providing the first SAE algorithm with theoretical recovery guarantees and practical effectiveness for LLM interpretation.

Deep Learning · Generative Models and Autoencoders

Sergei Kholkin, Ivan Butakov, Evgeny Burnaev, Nikita Gushchin, Aleksandr Korotin

Diffusion bridge models have recently become a powerful tool in the field of generative modeling. In this work, we leverage their power to address another important problem in machine learning and information theory, the estimation of the mutual information (MI) between two random variables. Neatly framing MI estimation as a domain transfer problem, we construct an unbiased estimator for data posing difficulties for conventional MI estimators. We showcase the performance of our estimator on three standard MI estimation benchmarks, i.e., low-dimensional, image-based and high MI, and on real-world data, i.e., protein language model embeddings.

Deep Learning · Theory

Jianzhe Wei, Siyu Chen, Jianliang He, Zhuoran Yang

Transformers have demonstrated remarkable in-context learning abilities, adapting to new tasks from just a few examples without parameter updates. However, theoretical understanding of this phenomenon typically assumes fixed dependency structures, while real-world sequences exhibit flexible, context-dependent relationships. We address this gap by investigating whether transformers can learn causal structures -- the underlying dependencies between sequence elements -- directly from in-context examples. We propose a novel framework using Markov chains with randomly sampled causal dependencies, where transformers must infer which tokens depend on which predecessors to make accurate predictions. Our key contributions are threefold: (1) We prove that a two-layer transformer with relative positional embeddings can implement Bayesian Model Averaging (BMA), the optimal statistical algorithm for causal structure inference; (2) Through extensive experiments and parameter-level analysis, we demonstrate that transformers trained on this task approximate BMA, with attention patterns directly reflecting the inferred causal structures; (3) We provide information-theoretic guarantees showing how transformers recover causal dependencies and extend our analysis to continuous dynamical systems, revealing fundamental differences in representational requirements. Our findings bridge the gap between empirical observations of in-context learning and theoretical understanding, showing that transformers can perform sophisticated statistical inference over structural uncertainty.

Applications · Chemistry and Drug Discovery

Heewoong Noh, Namkyeong Lee, Gyoung S. Na, Kibum Kim, Chanyoung Park

Spectral analysis provides crucial clues for the elucidation of unknown materials. Among various techniques, infrared spectroscopy (IR) plays an important role in laboratory settings due to its high accessibility and low cost. However, existing approaches often fail to reflect expert analytical processes and lack flexibility in incorporating diverse types of chemical knowledge, which is essential in real-world analytical scenarios. In this paper, we propose IR-Agent, a novel multi-agent framework for molecular structure elucidation from IR spectra. The framework is designed to emulate expert-driven IR analysis procedures and is inherently extensible. Each agent specializes in a specific aspect of IR interpretation, and their complementary roles enable integrated reasoning, thereby improving the overall accuracy of structure elucidation. Through extensive experiments, we demonstrate that IR-Agent not only improves baseline performance on experimental IR spectra but also shows strong adaptability to various forms of chemical information. The source code for IR-Agent is available at https://github.com/HeewoongNoh/IR-Agent.

General Machine Learning · Representation Learning

Ya Wen, Jixuan Cai, Qiyao Ma, Linyan Li, Xinhuan Chen, Chris Webster, Yulun Zhou

Representation learning of geospatial locations remains a core challenge in achieving general geospatial intelligence, with increasingly diverging philosophies and techniques. While Earth observation paradigms excel at depicting locations in their physical states, we propose that a location’s full characterization requires grounding in both its physical attributes and its internal human activity pattern, the latter being particularly crucial for understanding its human-centric functions. We present MoRA, a human-centric geospatial framework that leverages a mobility graph as its core backbone to fuse various data modalities, aiming to learn embeddings that represent the socio-economic context and functional role of a location. MoRA achieves this through the integration of spatial tokenization, GNNs, and asymmetric contrastive learning to align 100M+ POIs, massive remote sensing imagery, and structured demographic statistics with a billion-edge mobility graph, ensuring the three auxiliary modalities are interpreted through the lens of fundamental human dynamics. To rigorously evaluate the effectiveness of MoRA, we construct a benchmark dataset composed of 9 downstream prediction tasks across social and economic domains. Experiments show that MoRA, with four input modalities and a compact 128-dimensional representation space, achieves superior predictive performances than state-of-the-art models by an average of 12.9\%. Echoing LLM scaling laws, we further demonstrate the scaling behavior in geospatial representation learning. We open-source code and pretrained models at: https://github.com/ylzhouchris/MoRA.

Applications · Robotics

Jialiang Zhang, Haoran Geng, Yang You, Congyue Deng, Pieter Abbeel, Jitendra Malik, Leonidas Guibas

Understanding and predicting articulated actions is important in robot learning. However, common architectures such as MLPs and Transformers lack inductive biases that reflect the underlying kinematic structure of articulated systems. To this end, we propose the **Neural Rodrigues Operator**, a learnable generalization of the classical forward kinematics operation, designed to inject kinematics-aware inductive bias into neural computation. Building on this operator, we design the **Rodrigues Network (RodriNet)**, a novel neural architecture specialized for processing actions. We evaluate the expressivity of our network on two synthetic tasks on kinematic and motion prediction, showing significant improvements compared to standard backbones. We further demonstrate its effectiveness in two realistic applications: (i) imitation learning on robotic benchmarks with the Diffusion Policy, and (ii) single-image 3D hand reconstruction. Our results suggest that integrating structured kinematic priors into the network architecture improves action learning in various domains.

General Machine Learning · Transfer, Multitask and Meta-learning

Enneng Yang, Qun Yang, Peng Wang, Anke Tang, Guibing Guo, Xiaochun Cao, Li Shen

Model merging aims to integrate multiple independently fine-tuned expert models into a single model while preserving the knowledge of all experts. However, existing approaches mainly address parameter conflicts at the merging stage and overlook the role of the fine-tuning process, which often leads to significant post-merge performance degradation. To address this limitation, we propose a novel merging-aware optimizer (abbreviated as MergOPT) that injects principled merge-induced parameter shifts into the weight update steps so that the fine-tuned model exhibits a more stable loss landscape under subsequent merging operations. Specifically, we first formulate model merging as a distributionally robust optimization problem in the weight space: the parameters of other experts to be merged are viewed as adversarial merge-offsets, and fine-tuning adapts to the worst-case merging scenario. Building on this formulation, we analyze the distribution of parameter updates and the effects of merging hyperparameters, from which we derive a merging-guided feasible region for weight shifts. Finally, extensive experiments across four large language models (LLMs) and one vision model show that our approach consistently outperforms standard fine-tuning, yielding an average relative gain of 3.5\% and a maximum gain of 9.5\% across four merging strategies when merging seven experts.

Computer Vision · 3D Rendering & Reconstruction

Hanzhou Liu, Jia Huang, Mi Lu, Srikanth Saripalli, Peng Jiang

We present Stylos, a single-forward 3D Gaussian framework for 3D style transfer that operates on unposed content, from a single image to a multi-view collection, conditioned on a separate reference style image. Stylos synthesizes a stylized 3D Gaussian scene without per-scene optimization or precomputed poses, achieving geometry-aware, view-consistent stylization that generalizes to unseen categories, scenes, and styles. At its core, Stylos adopts a Transformer backbone with two pathways: geometry predictions retain self-attention to preserve geometric fidelity, while style is injected via cross-attention to enforce visual consistency across views. With the addition of a voxel-based 3D style loss that aligns aggregated scene features to style statistics, Stylos enforces view-consistent stylization while maintaining geometric coherence. Experiments across multiple datasets demonstrate that Stylos delivers high-quality zero-shot stylization, highlighting the effectiveness of the proposed style-content fusion block, the voxel-level style loss, and the scalability of our framework from single view to large-scale multi-view settings. Our codes are available at https://github.com/HanzhouLiu/Stylos.

Computer Vision · Vision Models & Multimodal

Jiwan Kim, Kibum Kim, Sangwoo Seo, Chanyoung Park

Recently, efficient Multimodal Large Language Models (MLLMs) have gained significant attention as a solution to their high computational complexity, making them more practical for real-world applications. In this regard, the knowledge distillation (KD) approach has emerged as a promising alternative, which transfers the rich visual and linguistic knowledge from a larger model (teacher) to a smaller model (student). However, we observe that existing KD methods struggle to effectively distill the teacher MLLM's rich visual perception abilities to the student, a challenge that has been largely overlooked in previous studies. Through a systematic analysis, we identify visual attention misalignment between student and teacher as the main cause of this issue. Based on this insight, we propose CompoDistill, a novel KD framework that explicitly aligns the student's visual attention with that of the teacher to enhance the student's visual perception abilities. Our extensive experiments show that CompoDistill significantly improves performance on compositional reasoning tasks that require visual perception abilities while maintaining strong performance on visual question answering tasks, as done in existing studies. Furthermore, CompoDistill demonstrates effectiveness with a more advanced backbone, highlighting its generalizability.

General Machine Learning · Everything Else

Wei Wang, Dong-Dong Wu, Ming Li, Jingxiong Zhang, Gang Niu, Masashi Sugiyama

Positive-unlabeled (PU) learning is a weakly supervised binary classification problem, in which the goal is to learn a binary classifier from only positive and unlabeled data, without access to negative data. In recent years, many PU learning algorithms have been developed to improve model performance. However, experimental settings are highly inconsistent, making it difficult to identify which algorithm performs better. In this paper, we propose the first PU learning benchmark to systematically compare PU learning algorithms. During our implementation, we identify subtle yet critical factors that affect the realistic and fair evaluation of PU learning algorithms. On the one hand, many PU learning algorithms rely on a validation set that includes negative data for model selection. This is unrealistic in traditional PU learning settings, where no negative data are available. To handle this problem, we systematically investigate model selection criteria for PU learning. On the other hand, PU learning involves different problem settings and corresponding solution families, i.e., the one-sample and two-sample settings. However, existing evaluation protocols are heavily biased towards the one-sample setting and neglect the significant difference between them. We identify the internal label shift problem of unlabeled training data for the one-sample setting and propose a simple yet effective calibration approach to ensure fair comparisons within and across families. We hope our framework will provide an accessible, realistic, and fair environment for evaluating PU learning algorithms in the future.

Computer Vision · Image and Video Generation

Łukasz Staniszewski, Łukasz Kuciński, Kamil Deja

Diffusion Models achieve state-of-the-art performance in generating new samples but lack a low-dimensional latent space that encodes the data into editable features. Inversion-based methods address this by reversing the denoising trajectory, transferring images to their approximated starting noise. In this work, we thoroughly analyze this procedure and focus on the relation between the initial noise, the generated samples, and their corresponding latent encodings obtained through the DDIM inversion. First, we show that latents exhibit structural patterns in the form of less diverse noise predicted for smooth image areas (e.g., plain sky). Through a series of analyses, we trace this issue to the first inversion steps, which fail to provide accurate and diverse noise. Consequently, the DDIM inversion space is notably less manipulative than the original noise. We show that prior inversion methods do not fully resolve this issue, but our simple fix, where we replace the first DDIM Inversion steps with a forward diffusion process, successfully decorrelates latent encodings and enables higher quality editions and interpolations.