NeurIPS 2025 论文列表

Photo retouching has become integral to contemporary visual storytelling, enabling users to capture aesthetics and express creativity. While professional tools such as Adobe Lightroom offer powerful capabilities, they demand substantial expertise and manual effort. In contrast, existing AI-based solutions provide automation but often suffer from limited adjustability and poor generalization, failing to meet diverse and personalized editing needs. To bridge this gap, we introduce JarvisArt, a multi-modal large language model (MLLM)-driven agent that understands user intent, mimics the reasoning process of professional artists, and intelligently coordinates over 200 retouching tools within Lightroom. JarvisArt undergoes a two-stage training process: an initial Chain-of-Thought supervised fine-tuning to establish basic reasoning and tool-use skills, followed by Group Relative Policy Optimization for Retouching (GRPO-R) to further enhance its decision-making and tool proficiency. We also propose the Agent-to-Lightroom Protocol to facilitate seamless integration with Lightroom. To evaluate performance, we develop MMArt-Bench, a novel benchmark constructed from real-world user edits. JarvisArt demonstrates user-friendly interaction, superior generalization, and fine-grained control over both global and local adjustments, paving a new avenue for intelligent photo retouching. Notably, it outperforms GPT-4o with a 60\% improvement in average pixel-level metrics on MMArt-Bench for content fidelity, while maintaining comparable instruction-following capabilities.

Despite the remarkable reasoning performance, eliciting the long chain-of-thought(CoT) ability in large language models(LLMs) typically requires costly reinforcement learning or supervised fine-tuning on high-quality distilled data. We investigate the internal mechanisms behind this capability and show that a small set of high-impact activations in the last few layers, greatly govern the long-form reasoning attributes, e.g. output length and self-reflection. Through simply amplifying these activations and adding ``wait'' tokens, the long CoT ability can be invoked without training, leading to significantly increased self-reflection rate and accuracy. In addition, we also find that the activation changes follow predictable trajectories, i.e. a sharp rise after special tokens and a subsequent exponential decay. Based on these insights, we introduce a general training-free activation control technique. It utilizes a few contrastive examples to identify the relevant activations, and then incorporates simple analytic functions to adjust their values at inference time to elicit long CoTs. Extensive experiments have verified the effectiveness of our methods in efficiently eliciting the long CoT ability of LLMs and improving the performance. Besides, we further propose a parameter-efficient fine-tuning method that trains only the last-layer activation amplification module and a few LoRA layers, outperforming LoRA on reasoning benchmarks with much fewer parameters. Our code and data will be fully public released.

Despite the impressive performance of large language models (LLMs), the process of endowing them with new capabilities---such as mathematical reasoning---remains largely empirical and opaque. A critical open question is whether reasoning abilities stem from the entire model, specific modules, or are merely artifacts of overfitting. In this work, we hypothesize that the reasoning capabilities in well-trained LLMs are primarily attributed to the output projection module (o_proj) in the Transformer’s multi-head self-attention (MHSA) module. To support this hypothesis, we introduce Stethoscope for Networks (SfN), a suite of diagnostic tools designed to probe and analyze the internal behaviors of LLMs. Using SfN, we provide both circumstantial and empirical evidence suggesting that o_proj plays a central role in enabling reasoning, whereas other modules contribute more to fluent dialogue. These findings offer a new perspective on LLM interpretability and open avenues for more targeted training strategies, potentially enabling more efficient and specialized LLMs.

Segmentation models can recognize a pre-defined set of objects in images. However, segmentation models capable of "reasoning" over complex user queries that implicitly refer to multiple objects of interest remain underexplored, especially in the geospatial domain. Recent advances in "reasoning segmentation"---generating segmentation masks from complex, implicit query text---demonstrate the potential of vision-language models (VLMs) to reason across an open domain of objects. Yet, our experiments reveal that these models struggle when applied to the unique challenges of remote-sensing imagery. To address this gap, we introduce a new dataset which consists of: GRES, a curated geospatial reasoning-segmentation dataset with 27,615 annotations across 9,205 images, and PreGRES, a collection of existing datasets to make up a large-scale multimodal pretraining corpus with over 1M question-answer pairs across 119,279 images. We propose an initial benchmark model, LISAt, a VLM for geospatial analysis that can describe complex remote-sensing scenes, answer detailed queries, and segment objects based on natural-language prompts. LISAt establishes a strong initial geospatial benchmark, outperforming prior foundation models such as RS-GPT4V by 10.04\% (BLEU-4) on visual description tasks and surpassing open-domain models on geospatial reasoning segmentation by 143.36\% (gIoU). Our model, dataset, and code are available on our project page: https://lisat-bair.github.io/LISAt/.

Recently, Large Foundation Models (LFMs), e.g., CLIP and GPT, have significantly advanced the Human-Object Interaction (HOI) detection, due to their superior generalization and transferability. Prior HOI detectors typically employ single- or multi-modal prompts to generate discriminative representations for HOIs from pretrained LFMs. However, such prompt-based approaches focus on transferring HOI-specific knowledge, but unexplore the potential reasoning capabilities of LFMs, which can provide informative context for ambiguous and open-world interaction recognition. In this paper, we propose InstructHOI, a novel method that leverages context-aware instructions to guide multi-modal reasoning for HOI detection. Specifically, to bridge knowledge gap and enhance reasoning abilities, we first perform HOI-domain fine-tuning on a pretrained multi-modal LFM, using a generated dataset with 140K interaction-reasoning image-text pairs. Then, we develop a Context-aware Instruction Generator (CIG) to guide interaction reasoning. Unlike traditional language-only instructions, CIG first mines visual interactive context at the human-object level, which is then fused with linguistic instructions, forming multi-modal reasoning guidance. Furthermore, an Interest Token Selector (ITS) is adopted to adaptively filter image tokens based on context-aware instructions, thereby aligning reasoning process with interaction regions. Extensive experiments on two public benchmarks demonstrate that our proposed method outperforms the state-of-the-art ones, under both supervised and zero-shot settings.

Long reasoning models (LRMs) such as OpenAI's o1 and DeepSeek's R1 have demonstrated remarkable abilities in complex reasoning tasks by scaling test-time compute and exhibiting human-like deep thinking. However, we identify a phenomenon we term underthinking, where LRMs frequently switch between different reasoning thoughts without sufficiently exploring promising paths to reach a correct solution. This behavior leads to inadequate depth of reasoning and decreased performance, particularly on challenging mathematical problems. To systematically analyze this issue, we conduct experiments on three challenging test sets and two representative open-source LRMs, revealing that frequent thought switching correlates with incorrect responses. We introduce a novel metric to quantify underthinking by measuring token efficiency in incorrect answers. To address underthinking, we propose a decoding strategy with thought switching penalty (Tip) that discourages premature transitions between thoughts, encouraging deeper exploration of each reasoning path. Experimental results demonstrate that our approach improves accuracy across challenging datasets without requiring model fine-tuning. Our findings contribute to understanding reasoning inefficiencies in LRMs and offer a practical solution to enhance their problem-solving capabilities. Our code is open-source and available at https://github.com/wangyuenlp/underthinking.

Emerging reasoning LLMs such as OpenAI-o1 and DeepSeek-R1 have achieved strong performance on complex reasoning tasks by generating long chain-of-thought (CoT) traces. However, these long CoTs result in increased token usage, leading to higher inference latency and memory consumption. As a result, balancing accuracy and reasoning efficiency has become essential for deploying reasoning LLMs in practical applications. Existing long-to-short (Long2Short) methods aim to reduce inference length but often sacrifice accuracy, revealing a need for an approach that maintains performance while lowering token costs. To address this efficiency-accuracy tradeoff, we propose TokenSqueeze, a novel Long2Short method that condenses reasoning paths while preserving performance and relying exclusively on self-generated data. First, to prevent performance degradation caused by excessive compression of reasoning depth, we propose to select self-generated samples whose reasoning depth is adaptively matched to the complexity of the problem. To further optimize the linguistic expression without altering the underlying reasoning paths, we introduce a distribution-aligned linguistic refinement method that enhances the clarity and conciseness of the reasoning path while preserving its logical integrity. Comprehensive experimental results demonstrated the effectiveness of TokenSqueeze in reducing token usage while maintaining accuracy. Notably, DeepSeek‑R1‑Distill‑Qwen‑7B fine-tuned by using our proposed method achieved a 50\% average token reduction while preserving accuracy on the MATH500 benchmark. TokenSqueeze exclusively utilizes the model's self-generated data, enabling efficient and high-fidelity reasoning without relying on manually curated short-answer datasets across diverse applications. Our code is available at \url{https://github.com/zhangyx1122/TokenSqueeze}.

Verifying the behavior of neural networks is necessary if developers are to confidently deploy them as parts of mission-critical systems. Toward this end, researchers have been actively developing a range of increasingly sophisticated and scalable neural network verifiers. However, scaling verification to large networks is challenging, at least in part due to the significant memory requirements of verification algorithms. In this paper, we propose an assume-guarantee compositional framework, CoVeNN, that is parameterized by an underlying verifier to generate a sequence of verification sub-problems to address this challenge. We present an iterative refinement-based strategy for computing assumptions that allow sub-problems to retain sufficient accuracy. An evaluation using 7 neural networks and a total of 140 property specifications demonstrates that CoVeNN can verify nearly 7 times more problems than state-of-the-art verifiers. CoVeNN is part of the NeuralSAT verification project: https://github.com/dynaroars/neuralsat.

Large language models (LLMs) have been widely adopted across diverse domains of software engineering, such as code generation, program repair, and vulnerability detection. These applications require understanding beyond surface-level code patterns: value propagation, control flow, and interdependence between program elements. However, existing benchmarks primarily evaluate end-to-end outcomes, such as whether code is correctly repaired or generated, leaving the models' ability of program semantic reasoning underexplored. This work presents CoRe, a high-quality, human-verified benchmark designed to evaluate LLMs on fundamental static analysis tasks. CoRe includes 12,553 task instances spanning data dependency, control dependency, and information flow across programs written in C/C++, Java, and Python. To ensure semantic diversity and reasoning complexity, we propose a semantics-aware diverse sampling strategy that selects targets and task instances based on structural coverage and dependency depth. We evaluate 10 state-of-the-art LLMs and show that, while they perform well at identifying dependencies, models still struggle with tasks that require deeper semantic understanding and multi-step reasoning. We further conduct qualitative analyses to uncover key challenges, such as complex control structures and backward dependency patterns, offering insights into improving LLMs’ code reasoning capabilities.

Compositional Zero-Shot Learning (CZSL) aims to recognize novel attribute-object compositions based on the knowledge learned from seen ones. Existing methods suffer from performance degradation caused by the distribution shift of label space at test time, which stems from the inclusion of unseen compositions recombined from attributes and objects. To overcome the challenge, we propose a novel approach that accumulates comprehensive knowledge in both textual and visual modalities from unsupervised data to update multimodal prototypes at test time. Building on this, we further design an adaptive update weight to control the degree of prototype adjustment, enabling the model to flexibly adapt to distribution shift during testing. Moreover, a dynamic priority queue is introduced that stores high-confidence images to acquire visual knowledge from historical images for inference. Considering the semantic consistency of multimodal knowledge, we align textual and visual prototypes by multimodal collaborative representation learning. Extensive experiments indicate that our approach achieves state-of-the-art performance on four benchmark datasets under both closed-world and open-world settings. Code will be available at [https://github.com/xud-yan/TOMCAT](https://github.com/xud-yan/TOMCAT).

Recent masked diffusion models (MDMs) have shown competitive performance compared to autoregressive models (ARMs) for language modeling. While most literature has focused on performance enhancing sampling procedures, efficient sampling from MDMs has been scarcely explored. We make the observation that often a given sequence of partially masked tokens determines the values of multiple unknown tokens deterministically, meaning that a single prediction of a masked model holds additional information unused by standard sampling procedures. Based on this observation, we introduce *EB-Sampler*, a simple drop-in replacement for existing samplers, utilizing an **E**ntropy **B**ounded unmasking procedure that dynamically unmasks multiple tokens in one function evaluation with predefined approximate error tolerance. We formulate the EB-Sampler as part of a broad family of adaptive samplers for which we provide an error analysis that motivates our algorithmic choices. EB-Sampler accelerates sampling from current state of the art MDMs by roughly 2-3x on standard coding and math reasoning benchmarks without loss in performance. We also validate the same procedure works well on smaller reasoning tasks including maze navigation and sudoku, tasks ARMs often struggle with.

Large language models (LLMs) have demonstrated remarkable reasoning capabilities through test-time scaling approaches, particularly when fine-tuned with chain-of-thought (CoT) data distilled from more powerful large reasoning models (LRMs). However, these reasoning chains often contain verbose elements that mirror human problem-solving, categorized as progressive reasoning (the essential solution development path) and functional elements (verification processes, alternative solution approaches, and error corrections). While progressive reasoning is crucial, the functional elements significantly increase computational demands during test-time inference. We introduce PIR (Perplexity-based Importance Refinement), a principled framework that quantitatively evaluates the importance of each reasoning step based on its impact on answer prediction confidence. PIR systematically identifies and selectively prunes only low-importance functional steps while preserving all progressive reasoning components, creating optimized training data that maintains the integrity of the core solution path while reducing verbosity. Models fine-tuned on PIR-optimized data exhibit superior test-time scaling properties, generating more concise reasoning chains while achieving improved accuracy (+0.9\% to +6.6\%) with significantly reduced token usage (-3\% to -41\%) across challenging reasoning benchmarks (AIME, AMC, and GPQA Diamond). Our approach demonstrates strong generalizability across different model sizes, data sources, and token budgets, offering a practical solution for deploying reasoning-capable LLMs in scenarios where efficient test-time scaling, response time, and computational efficiency are valuable constraints. Code and dataset are available at the [LIMOPro GitHub repository.](https://github.com/GAIR-NLP/LIMOPro)

Collecting ground-truth rewards or human demonstrations for multi-step reasoning tasks is often prohibitively expensive, especially in interactive domains such as web tasks. We introduce Self-Taught Lookahead (STL), a reward-free framework that improves language model–based value functions by reasoning explicitly about state transitions. STL can be viewed as a chain-of-thought analogue of the value iteration algorithm: instead of regressing directly on numeric values, a value LLM is trained to simulate a step of lookahead in natural language—predicting the next action, resulting state, and rationale for its value. This process refines value estimates without any labeled data. The self-supervised procedure yields more accurate state-value predictions, which in turn enable lightweight search algorithms to expand fewer states while maintaining strong performance. Empirically, STL-trained value models built on moderately sized (8B-parameter) open-weight LLMs boost web agent success rates by over 39%, achieving performance comparable to proprietary models. STL also generalizes to multi-hop question answering and math puzzles. Overall, STL enables small open-source models to guide efficient search, reducing inference costs by integrating explicit reasoning with value learning.

We present scPilot, the first systematic framework to practice \textit{omics-native reasoning}: a large language model (LLM) converses in natural language while directly inspecting single-cell RNA-seq data and on-demand bioinformatics tools. scPilot converts core single-cell analyses, i.e., cell-type annotation, developmental-trajectory reconstruction, and transcription-factor targeting, into step-by-step reasoning problems that the model must solve, justify, and, when needed, revise with new evidence. To measure progress, we release \scbench, a suite of 9 expertly curated datasets and graders that faithfully evaluate the omics-native reasoning capability of scPilot w.r.t various LLMs. Experiments with o1 show that \textit{iterative} omics-native reasoning lifts average accuracy by 11\% for cell-type annotation and Gemini 2.5 Pro cuts trajectory graph-edit distance by 30\% versus one-shot prompting, while generating transparent reasoning traces that explain marker gene ambiguity and regulatory logic. By grounding LLMs in raw omics data, scPilot enables auditable, interpretable, and diagnostically informative single-cell analyses.

Reinforcement learning (RL) is a critical component of large language model (LLM) post-training. However, on-policy algorithms used for post-training are not naturally robust to a diversified content of experience replay buffers, which asynchronous off-policy actors can efficiently populate in parallel to training. We propose efficiently learning on such off-policy data via Trajectory Balance with Asynchrony (TBA), an approach to asynchronous RL for LLMs that leverages the principled off-policy TB objective. On math, preference-tuning, and automated red-teaming tasks, we post-train models ranging from Pythia 410M to Qwen 2.5 7B, finding TBA offers speed and performance boosts over strong baselines like Online DPO and Dr. GRPO. Beyond TBA's performance benefits (high accuracy even as asynchrony grows) and speedups ($4\times$ or more), we show its reward- and recency-prioritizing sampling enable further gains as data generation is scaled. Our code is available at https://github.com/bbartoldson/TBA.

Understanding perspective is fundamental to human visual perception, yet the extent to which multimodal large language models (MLLMs) internalize perspective geometry remains unclear. We introduce MMPerspective, the first benchmark specifically designed to systematically evaluate MLLMs' understanding of perspective through 10 carefully crafted tasks across three complementary dimensions: Perspective Perception, Reasoning, and Robustness. Our benchmark comprises 2,711 real-world and synthetic image instances with 5,083 question-answer pairs that probe key capabilities, such as vanishing point perception and counting, perspective type reasoning, line relationship understanding in 3D space, invariance to perspective-preserving transformations, etc. Through a comprehensive evaluation of 43 state-of-the-art MLLMs, we uncover significant limitations: while models demonstrate competence on surface-level perceptual tasks, they struggle with compositional reasoning and maintaining spatial consistency under perturbations. Our analysis further reveals intriguing patterns between model architecture, scale, and perspective capabilities, highlighting both robustness bottlenecks and the benefits of chain-of-thought prompting. MMPerspective establishes a valuable testbed for diagnosing and advancing spatial understanding in vision-language systems. Resources are available at https://yunlong10.github.io/MMPerspective/

Large language models (LLMs) excel at complex reasoning tasks but remain computationally expensive, limiting their practical deployment. To address this, recent works have focused on distilling reasoning capabilities into smaller language models (sLMs) using chain-of-thought (CoT) traces from teacher LLMs. However, this approach struggles in scenarios requiring rare factual knowledge or precise computation, where sLMs often hallucinate due to limited capability. In this work, we propose Agent Distillation, a framework for transferring not only reasoning capability but full task-solving behavior from LLM-based agents into sLMs with retrieval and code tools. We improve agent distillation along two complementary axes: (1) we introduce a prompting method called first-thought prefix to enhance the quality of teacher-generated trajectories; and (2) we propose a self-consistent action generation for improving test-time robustness of small agents. We evaluate our method on eight reasoning tasks across factual and mathematical domains, covering both in-domain and out-of-domain generalization. Our results show that sLMs as small as 0.5B, 1.5B, 3B parameters can achieve performance competitive with next-tier larger 1.5B, 3B, 7B models fine-tuned using CoT distillation, demonstrating the potential of agent distillation for building practical, tool-using small agents.

LLMs often need effective configurations, like temperature and reasoning steps, to handle tasks requiring sophisticated reasoning and problem-solving, ranging from joke generation to mathematical reasoning. Existing prompting approaches usually adopt general-purpose, fixed configurations that work “well enough” across tasks but seldom achieve task-specific optimality. To address this gap, we introduce AdaReasoner, an LLM-agnostic plugin designed for any LLM to automate adaptive reasoning configurations for tasks requiring different types of thinking. AdaReasoner is trained using a reinforcement learning (RL) framework, combining a factorized action space with a targeted exploration strategy, along with a pretrained reward model to optimize the policy model for reasoning configurations with only a few-shot guide. AdaReasoner is backed by theoretical guarantees and experiments of fast convergence and a sublinear policy gap. Across six different LLMs and a variety of reasoning tasks, it consistently outperforms standard baselines, preserves out-of-distribution robustness, and yield gains on knowledge-intensive tasks through tailored prompts.

State Space Models (SSMs) have emerged as efficient alternatives to Vision Transformers (ViTs), with VMamba standing out as a pioneering architecture designed for vision tasks. However, their generalization performance degrades significantly under distribution shifts. To address this limitation, we propose TRUST (Test-Time Refinement using Uncertainty-Guided SSM Traverses), a novel test-time adaptation (TTA) method that leverages diverse traversal permutations to generate multiple causal perspectives of the input image. Model predictions serve as pseudo-labels to guide updates of the Mamba-specific parameters, and the adapted weights are averaged to integrate the learned information across traversal scans. Altogether, TRUST is the first approach that explicitly leverages the unique architectural properties of SSMs for adaptation. Experiments on seven benchmarks show that TRUST consistently improves robustness and outperforms existing TTA methods.