论文广场 - AcademicHub

01.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2504.14582

NTIRE 2025 Challenge on Image Super-Resolution (x4): Methods and Results

作者:

Zheng Chen ↗Kai Liu ↗Jue Gong ↗Jingkai Wang ↗Lei Sun ↗Zongwei Wu ↗Radu Timofte ↗Yulun Zhang ↗Xiangyu Kong ↗Xiaoxuan Yu ↗Hyunhee Park ↗Suejin Han ↗…

This paper presents the NTIRE 2025 image super-resolution ($\times$4) challenge, one of the associated competitions of the 10th NTIRE Workshop at CVPR 2025. The challenge aims to recover high-resolution (HR) images from low-resolution (LR) counterparts generated through bicubic downsampling with a $\times$4 scaling factor. The objective is to develop effective network designs or solutions that achieve state-of-the-art SR performance. To reflect the dual objectives of image SR research, the challenge includes two sub-tracks: (1) a restoration track, emphasizes pixel-wise accuracy and ranks submissions based on PSNR; (2) a perceptual track, focuses on visual realism and ranks results by a perceptual score. A total of 286 participants registered for the competition, with 25 teams submitting valid entries. This report summarizes the challenge design, datasets, evaluation protocol, the main results, and methods of each team. The challenge serves as a benchmark to advance the state of the art and foster progress in image SR.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.15197

StarOR: Synergizing Tree Search and Test-Time Reinforcement Learning for Optimization Modeling

作者:

Jiajun Li ↗Yu Ding ↗Shisi Guan ↗Ran Hou ↗Wanyuan Wang ↗

arXiv:2606.15197v1 Announce Type: cross Abstract: Optimization modeling is inherently hierarchical, requiring a precise sequence of symbolic commitments. Traditional learning-based automated optimization modeling methods improve modeling policies through large-scale annotated or curated training data, but are costly to adapt to new problem distributions. Meanwhile, one-shot generation remains brittle in hierarchical modeling, where early symbolic errors can propagate into invalid formulations. Test-time scaling offers a promising alternative by enabling structural exploration with additional instance-level computation; however, existing search-based methods typically rely on a fixed policy, causing repeated rollouts to inherit similar modeling biases and providing limited credit assignment for intermediate decisions. To address these limitations, we propose StarOR, a synergistic search-and-adaptation framework that couples MCTS with Test-Time Reinforcement Learning for optimization modeling. StarOR decomposes the modeling process into four stages and updates a transient LoRA adapter via GRPO at each non-terminal node. By using MCTS-generated siblings as local comparison sets, StarOR transforms search-time exploration into instance-specific policy refinement. Moreover, an unsupervised multi-faceted reward system provides fine-grained feedback for intermediate formulation decisions without ground-truth labels. Experiments across five optimization benchmarks show that StarOR achieves state-of-the-art performance even with a 4B backbone, outperforming existing methods and the frontier LLMs.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2508.04266

ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents

作者:

Jiangyuan Wang ↗Kejun Xiao ↗Qi Sun ↗Huaipeng Zhao ↗Tao Luo ↗Jian Dong Zhang ↗Xiaoyi Zeng ↗

Existing benchmarks in e-commerce primarily focus on basic user intents, such as finding or purchasing products. However, real-world users often pursue more complex goals, such as applying vouchers, managing budgets, and finding multi-products seller. To bridge this gap, we propose ShoppingBench, a novel end-to-end shopping benchmark designed to encompass increasingly challenging levels of grounded intent. Specifically, we propose a scalable framework to simulate user instructions based on various intents derived from sampled real-world products. To facilitate consistent and reliable evaluations, we provide a large-scale shopping sandbox that serves as an interactive simulated environment, incorporating over 2.5 million real-world products. Experimental results demonstrate that even state-of-the-art language agents (such as GPT-4.1) achieve absolute success rates under 50% on our benchmark tasks, highlighting the significant challenges posed by our ShoppingBench. In addition, we propose a trajectory distillation strategy and leverage supervised fine-tuning, along with reinforcement learning on synthetic trajectories, to distill the capabilities of a large language agent into a smaller one. As a result, our trained agent achieves competitive performance compared to GPT-4.1.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.18974

Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning

作者:

Pengyu Li ↗Zhitao Gao ↗Lingling Zhang ↗Muye Huang ↗Yuanming Li ↗Fangzhi Xu ↗Jun Liu ↗

Unified multimodal models (UMMs) interleave generated ''visual thoughts'' (VTs) with text reasoning to improve spatial tasks. This incurs roughly an order-of-magnitude inference cost from multi-step diffusion. We find this cost yields limited direct benefit. On ThinkMorph, removing or noising VTs barely changes accuracy across nine benchmarks. Once rendered, attention concentrates on the VT regardless of content. Yet a KL diagnostic shows that conditioning on a privileged VT trace shifts the model's completion distribution. This suggests the generation pathway encodes useful reasoning beyond the rendered pixels. Motivated by this gap, we propose Visual On-Policy Self-Distillation(Visual-OPSD). Teacher and student share identical weights but differ in context: the teacher sees privileged VTs while the student sees only the question. Token-level JSD distillation on on-policy student trajectories transfers the teacher's reasoning to a text-only student. Across nine benchmarks, Visual-OPSD improves over its generative teacher by $+3.40$pp with $14.3\times$ speedup (10.0s vs. 142.8s per sample) and outperforms same-scale VLMs by $+63.83$pp on VSP. A Gaussian-noise control ($+0.40$pp vs. $+10.28$pp for real VTs) and $58.4\%$ closure of the KL gap confirm that gains come from the semantic content of the generation pathway.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2409.18909

Best Arm Identification with Minimal Regret

作者:

Junwen Yang ↗Vincent Y. F. Tan ↗Tianyuan Jin ↗

arXiv:2409.18909v2 Announce Type: replace Abstract: Motivated by real-world applications that necessitate responsible experimentation, we introduce the problem of best arm identification (BAI) with minimal regret. This variant of the multi-armed bandit problem elegantly amalgamates two of its most ubiquitous objectives: regret minimization and BAI. More precisely, the agent's goal is to identify the best arm with a prescribed confidence level $\delta$, while minimizing the cumulative regret up to the stopping time. Focusing on single-parameter exponential families of distributions, we leverage information-theoretic techniques to establish an instance-dependent lower bound on the expected cumulative regret. Moreover, we present an impossibility result that underscores the tension between cumulative regret and sample complexity in fixed-confidence BAI. Complementarily, we design and analyze the Double KL-UCB algorithm, which achieves asymptotic optimality as the confidence level tends to zero. Notably, this algorithm employs two distinct confidence bounds to guide arm selection in a randomized manner. Our findings elucidate a fresh perspective on the inherent connections between regret minimization and BAI.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2601.13591

DSAEval: Evaluating Data Science Agents on a Wide Range of Real-World Data Science Problems

作者:

Maojun Sun ↗Yifei Xie ↗Yue Wu ↗Ruijian Han ↗Binyan Jiang ↗Defeng Sun ↗Yancheng Yuan ↗Jian Huang ↗

Recent LLM-based data agents aim to automate data science tasks ranging from data analysis to deep learning. However, the open-ended nature of real-world data science problems, which often span multiple taxonomies and lack standard answers, poses a significant challenge for evaluation. To address this, we introduce DSAEval, a benchmark comprising 641 real-world data science problems grounded in 285 diverse datasets, covering both structured and unstructured data (e.g., image and text). DSAEval incorporates three distinctive features: (1) Multimodal Environment Perception, which enables agents to interpret observations from multiple modalities, including text and vision; (2) Multi-Query Interactions, which mirror the iterative and cumulative nature of real-world data science projects; and (3) Multi-Dimensional Evaluation, which provides a holistic assessment across reasoning, code, and results. We systematically evaluate 13 recent advanced agentic LLMs using DSAEval. Our results show that Claude-Sonnet-4.5 achieves the strongest overall performance, MiMo-V2-Pro and GPT-5.2 lead in duration and step efficiency, respectively, and MiMo-V2-Flash is the most cost-effective. We further demonstrate that multimodal perception consistently improves performance on vision-related tasks, with gains ranging from 2.04\% to 11.30\%. Overall, while current data science agents perform well on structured data and routine data analysis workflows, substantial challenges remain in unstructured domains. Finally, we offer critical insights and outline future research directions.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.18566

Multi-Modal Hyper-Graph Fusion for Low-Light Crowd Counting

作者:

Hao-Yuan Ma ↗Li Zhang ↗Yushi Qiu ↗Jie Gao ↗Yan Zhang ↗Bangjun Wang ↗

Crowd counting is a fundamental task in computer vision. However, crowd counting in low-light environments remains largely underexplored, despite its practical importance in the real world. Existing methods mainly focus on well-lit scenes or rely on single-modality Red-Green-Blue (RGB) representations, which often become unreliable under extreme darkness and complex non-uniform illumination. To handle this problem, we construct three new low-light crowd counting benchmarks, which consist of two synthetic datasets, SHA\_Dark and SHB\_Dark, and a real-world benchmark LC-Crowd (Low-light Crowd Dataset). Inspired by Retinex-based physical modeling, we introduce depth and Canny edge cues as complementary geometric and structural priors to enhance the intrinsic reflectance representation under low-light conditions. We propose a Multi-Modal Hyper-Graph Fusion module, which formulates RGB appearance, depth geometry, and edge structure cues as nodes in a unified hyper-graph and explicitly captures their high-order complementary relationships via dynamic hyperedge construction and message passing. Furthermore, to adaptively allocate computation in dense prediction, we propose a Deformable Rectangular Sparse Attention (DRSA) module, which concentrates computation on informative regions through anchor-aware estimation and adaptive rectangular window modeling. Based on these designs, we develop a unified Low-Light Counting Network (LCNet) for robust low-light crowd counting. Extensive experiments on three benchmarks demonstrate that the proposed method achieves the best overall performance against existing state-of-the-art (SOTA) methods. The code is in the supplementary material. The datasets will be made public upon acceptance.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2603.01761

Position: Modular Memory is the Key to Continual Learning Agents

作者:

arXiv:2603.01761v2 Announce Type: replace-cross Abstract: Foundation models have transformed machine learning through large-scale pretraining and increased test-time compute. Despite surpassing human performance in several domains, these models remain fundamentally limited in continuous operation, experience accumulation, and personalization, capabilities that are central to adaptive intelligence. While continual learning research has long targeted these goals, its historical focus on in-weight learning (IWL), i.e., updating a single model's parameters to absorb new knowledge, has rendered catastrophic forgetting a persistent challenge. Our position is that combining the strengths of In-Weight Learning (IWL) and the newly emerged capabilities of In-Context Learning (ICL) through the design of modular memory is the missing piece for continual adaptation at scale. We outline a conceptual framework for modular memory-centric architectures that leverage ICL for rapid adaptation and knowledge accumulation, and IWL for stable updates to model capabilities, charting a practical roadmap toward continually learning agents.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.09669

SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks

作者:

Hongcheng Gao ↗Hailong Qu ↗Jingyi Tang ↗Jiahao Wang ↗Zihao Huang ↗Hengkang Qiao ↗Shihong Huang ↗Junming Yang ↗Yi Li ↗Hongyixuan Yuan ↗Wenjie Li ↗Bohan Zeng ↗…

Spatial reasoning is a foundational capability for multimodal large language models (MLLMs) to perceive and operate within the physical world. However, existing benchmarks predominantly rely on passive evaluation (e.g., static VQA) or simulator-specific pipelines, failing to assess general interactive spatial understanding. We introduce SpatialWorld, a unified benchmark designed specifically for evaluating the interactive spatial understanding of multimodal agents in complex real-world tasks. Integrating eight heterogeneous simulation backends under a shared, simulator-agnostic protocol, SpatialWorld features 760 human-annotated tasks across diverse domains (e.g., household routines, travel, social collaboration). Agents must solve tasks under vision-only partial observability, actively gathering egocentric visual evidence and expressing decisions via a unified, text-based action interface native to MLLMs. For reliable evaluation, each task includes a human-validated initial state, a reference trajectory, and a terminal-state verifier. Evaluating 15 advanced agents reveals that robust spatial task solving remains challenging: the strongest model, GPT-5, achieves an average task success rate (TSR) of only 17.4%, while the leading open-source model, Qwen-3.5, reaches 14.1%. Further analysis exposes a clear mismatch between task success and execution efficiency, alongside substantial domain-specific performance variations. These bottlenecks in active exploration and long-horizon planning position SpatialWorld as a rigorous testbed for future spatial agents.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.CV) 2026-06-12 DOI: arXiv:2606.07436

Skill-3D: Evolving Scene-Aware Skills for Agentic 3D Spatial Reasoning

作者:

Haoyuan Li ↗Zhengdong Hu ↗Jun Wang ↗Hehe Fan ↗Yi Yang ↗

This paper explores agentic 3D spatial understanding, i.e., MLLM agents performing 3D reasoning through tool use. Existing methods often misuse tools and exhibit biased tool preferences under 3D scenarios, leaving the agentic paradigm with only marginal gains over non-agentic strategies. We reveal that 3D spatial reasoning tasks are heterogeneous across scenes, while these agents apply a uniform tool-use strategy to all scenes rather than selecting tools according to the specific scene and task. To address this, we propose Skill-3D, a framework that learns self-evolving scene-aware skills. Specifically, Skill-3D identifies the task scene and records the agent's tool-use trajectory into a Scene Memory, where successful trajectories from similar scenes are aggregated and distilled into a reusable scene-aware skill, with failed ones attached to the skill as lessons. During training, once a similar scene recurs, the corresponding skill is injected to guide the agent, producing new trajectories whose successes and failures further refine the skill, forming a loop in which the memory and the skill library co-evolve. Experiments show that Skill-3D substantially improves tool utilization in 3D spatial reasoning (from 39% to 78% on VSI-Bench), driving the agent toward correct and sufficient tool use. For instance, it improves Gemini-3-Flash by 67% on MMSI-Bench. Furthermore, we conduct agentic post-training over skill-guided trajectories, which boosts Qwen3-VL-8B by 60% on VSI-Bench.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.02800

Cosmos 3: Omnimodal World Models for Physical AI

作者:

NVIDIA ↗Aditi ↗Niket Agarwal ↗Arslan Ali ↗Jon Allen ↗Martin Antolini ↗Adeline Aubame ↗Alisson Azzolini ↗Junjie Bai ↗Maciej Bala ↗Yogesh Balaji ↗Josh Bapst ↗…

We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. By supporting highly flexible input-output configurations, Cosmos 3 seamlessly unifies critical modalities for Physical AI – effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework. Our evaluation demonstrates that Cosmos 3 establishes a new state-of-the-art across a diverse suite of understanding and generation tasks, demonstrating omnimodal world models as scalable, general-purpose backbones for embodied agents. Our post-trained Cosmos 3 models were ranked as the best open-source Text-to-Image and Image-to-Video models by Artificial Analysis, and the best policy model by RoboArena at the time the technical report was written. To accelerate open research and deployment in Physical AI, we make our code, model checkpoints, curated synthetic datasets, and evaluation benchmark available under the Linux Foundation's OpenMDW-1.1 License at https://github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3. The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos3.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.LG) 2026-06-19 DOI: arXiv:2606.19416

MortarBench: Evaluating Mortgage Loan Origination Agents

作者:

Matthew Toles ↗Yunan Lu ↗Manav Munjal ↗Bojun Liu ↗Yuanhao Deng ↗Stephanie Selig ↗Derek Rindner ↗Cheng Li ↗Zhou Yu ↗

arXiv:2606.19416v1 Announce Type: new Abstract: Loan origination is the process by which a lender creates a new loan, from application and underwriting through approval and funding. This process serves a critical role in evaluating the eligibility and level of risk posed by an applicant. Recently, firms have begun using mortgage loan agents to augment human loan officers, despite a lack of any public benchmark. To fill this gap, we present MortarBench, a loan origination agent benchmark. MortarBench uses a financial data synthesis and mutation pipeline to generate examples with broad edge case coverage that match real-world distributions and questions. We find that state-of-the-art large language models (LLMs) perform poorly, with closed-source models achieving at most 77.1\% exact match accuracy. We also discover systematic biases in LLM perception of foreignness related to non-English names. Noting these weaknesses, we introduce CRIT, a confidence calibration framework. Our method increases accuracy to 80.5\% while improving risk management steering and reducing bias.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.CV) 2026-06-15 DOI: arXiv:2606.13872

Avatar V: Scaling Video-Reference Avatar Video Generation

作者:

Benjamin Liang ↗Ce Chen ↗Desmond Lin ↗Ivan Somov ↗Jiajun Zhao ↗Jiewei Yuan ↗Jingfeng Zhang ↗Junhao Huang ↗Nik Nolte ↗Pedram Haqiqi ↗Penghan Wang ↗Rong Yan ↗…

Generating avatar videos that are not merely visually similar to a target individual but behaviorally recognizable, faithfully reproducing their talking rhythm, gestural tendencies, and expression dynamics, remains an open challenge. Existing methods predominantly condition on single static images, which provide insufficient identity information and cannot capture dynamic motion traits, while standard pixel-level objectives underserve the perceptually critical facial regions that determine avatar fidelity. We present Avatar V, a production-scale framework that addresses these limitations through video-reference-conditioned identity modeling. Rather than compressing identity into fixed-size embeddings, the model conditions directly on the full token sequence of a reference video, learning to reproduce both static identity attributes (facial geometry, skin texture) and dynamic behavioral patterns (talking rhythm, micro-expressions) through attention over the reference context. We introduce Sparse Reference Attention, an asymmetric mechanism achieving linear-complexity conditioning on arbitrarily long references; a motion representation stream enabling closed-loop talking style transfer; and an identity-aware super-resolution refiner inheriting the full reference conditioning. These are supported by a data engine curating 100M+ training clips from 50M raw videos, and a five-stage training pipeline with flow matching pre-training, personality fine-tuning, two-phase distillation (>10x acceleration), and RLHF alignment, deployed across thousands of GPUs. Avatar V generates 1080p videos of unlimited duration, achieving state-of-the-art identity preservation, lip synchronization, and generation quality on our cross-scene benchmark, consistently outperforming leading systems including Seedance 2.0, Kling O3 Pro, Veo 3.1, and OmniHuman 1.5 in both automated metrics and human evaluation.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.CV) 2026-06-19 DOI: arXiv:2606.20455

PCFootprint: A Large-Scale Dataset and Benchmark for Vectorized Building Footprint Extraction from Aerial LiDAR Point Clouds

作者:

Haoyuan Shen ↗Kuihao Wang ↗Ruisheng Wang ↗Yujun Liu ↗

Building footprint extraction is a fundamental task in photogrammetry, remote sensing, and computer vision. Recent image-based methods have achieved remarkable progress in extracting vectorized footprints from high-resolution optical imagery. However, optical imagery inherently susceptible to occlusions, perspective distortions, and residual relief displacement, yielding incomplete or misaligned footprint extraction. Furthermore, the lack of explicit elevation information limits its direct applicability to Level of Detail building modeling. In this paper, we present PCFootprint, the first large-scale public dataset for footprint extraction from airborne laser scanning point clouds. PCFootprint comprises \num{33000} tiles derived from the Estonian Land and Spatial Development Board, covering diverse urban and rural landscapes. Each tile spans \qtyproduct{128 x 128}{\m} with systematically aligned vectorized footprints aligned to point clouds. The dataset includes a \num{3000} tiles cross-domain test set for evaluating generalization across geographic regions. We establish comprehensive benchmarks by evaluating mainstream methods. Experimental results reveal significant challenges including high intra-class variance, data imbalance, and noise across complex geospatial environments. We believe PCFootprint will advance future research in building modeling, urban scene understanding, and geospatial analysis. The PCFootprint dataset is publicly available at \url{https://huggingface.co/datasets/Haoyuan-Shen/PCFootprint}.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15007

Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

作者:

We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 trillion text tokens, then extended the context length to 1M tokens, and post-trained using Supervised Fine Tuning (SFT), Reinforcement Learning (RL), and Multi-teacher On-Policy Distillation (MOPD). Nemotron 3 Ultra is our most capable model yet, employing multiple key technologies - LatentMoE, Multi Token Prediction (MTP), NVFP4 pre-training, multi-environment RLVR, MOPD, and reasoning budget control. Nemotron 3 Ultra achieves up to ~6x higher inference throughput as compared to state-of-the-art publicly available LLMs while attaining on-par accuracy. The state-of-the-art accuracy, high inference throughput, and 1M token context length make Nemotron 3 Ultra ideal for long-running autonomous agentic tasks. We open-source the base, post-trained, and quantized checkpoints, along with the training data and recipe on HuggingFace.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2511.18960

AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention

作者:

Lei Xiao ↗Jifeng Li ↗Juntao Gao ↗Feiyang Ye ↗Yan Jin ↗Jingjing Qian ↗Jing Zhang ↗Yong Wu ↗Xiaoyuan Yu ↗

Vision-Language-Action (VLA) models have shown remarkable progress in embodied tasks recently, but most methods process visual observations independently at each timestep. This history-agnostic design treats robot manipulation as a Markov Decision Process, even though real-world robotic control is inherently partially observable and requires reasoning over past interactions. To address this mismatch, we reformulate VLA policy learning from a Partially Observable Markov Decision Process perspective and propose AVA-VLA, a framework that conditions action generation on a recurrent state that serves as a neural approximation to the agent's belief over task history. Built on this recurrent state, we introduce Active Visual Attention (AVA), which dynamically reweights visual tokens in the current observation to focus on regions most relevant given both the instruction and execution history. Extensive experiments show that AVA-VLA achieves state-of-the-art performance on standard robotic benchmarks, including LIBERO and CALVIN, and transfers effectively to real-world dual-arm manipulation tasks. These results demonstrate the effectiveness of temporally grounded active visual processing for improving VLA performance in robotic sequential decision-making. The project page is available at https://liauto-dsr.github.io/AVA-VLA-Page.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2602.12279

UniT: Unified Multimodal Chain-of-Thought Test-time Scaling

作者:

Leon Liangyu Chen ↗Haoyu Ma ↗Zhipeng Fan ↗Ziqi Huang ↗Animesh Sinha ↗Xiaoliang Dai ↗Jialiang Wang ↗Zecheng He ↗Jianwei Yang ↗Chunyuan Li ↗Junzhe Sun ↗Chu Wang ↗…

Unified models can handle both multimodal understanding and generation within a single architecture, yet they typically operate in a single pass without iteratively refining their outputs. Many multimodal tasks, especially those involving complex spatial compositions, multiple interacting objects, or evolving instructions, require decomposing instructions, verifying intermediate results, and making iterative corrections. While test-time scaling (TTS) has demonstrated that allocating additional inference compute for iterative reasoning substantially improves language model performance, extending this paradigm to unified multimodal models remains an open challenge. We introduce UniT, a framework for multimodal chain-of-thought test-time scaling that enables a single unified model to reason, verify, and refine across multiple rounds. UniT combines agentic data synthesis, unified model training, and flexible test-time inference to elicit cognitive behaviors including verification, subgoal decomposition, and content memory. Our key findings are: (1) unified models trained on short reasoning trajectories generalize to longer inference chains at test time; (2) sequential chain-of-thought reasoning provides a more scalable and compute-efficient TTS strategy than parallel sampling; (3) training on generation and editing trajectories improves out-of-distribution visual reasoning. These results establish multimodal test-time scaling as an effective paradigm for advancing both generation and understanding in unified models.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15320

Conditional Multi-Event Temporal Grounding in Long-Form Video

作者:

Yuanhao Zou ↗Arthad Kulkarni ↗Lucas Tonanez ↗Lincoln Spencer ↗Guangyu Sun ↗Tianxingjian Ding ↗Andong Deng ↗Yi Li ↗Shuangjun Liu ↗Yuan Li ↗Dashan Gao ↗Ning Bi ↗…

Multimodal large language models have made rapid progress in video temporal grounding, yet real-world applications routinely require localizing every event that satisfies compositional temporal and spatial conditions. Existing benchmarks fall short: they localize only a single moment per query, count without temporal conditions, or treat grounding and counting as disjoint tasks. We introduce CoMET-Bench for Conditional Multi-Event Temporal Grounding in long-form video, comprising 2789 queries over 600 videos averaging 33.8 minutes across five real-world domains, with each query composed from 4 temporal conditions, 3 spatial conditions, and a dedicated negative-query subset. We further propose a unified evaluation protocol jointly measuring counting, grounding, and negative-query recognition, including a new Rejection-F1 metric that prevents trivial gaming by lazy "always-empty" models. Benchmarking a broad suite of MLLMs, agent-based, and grounding-specialized methods reveals that existing approaches remain far from solving this task. Building on these findings, we propose CoMET-Agent, a training-free agentic framework that reformulates the task as structured search-and-aggregate, improving F1@0.5 by 6.1% over GPT-5 purely through structural reasoning. Failure analysis further surfaces three open directions: fine-grained entity tracking, position-uniform retrieval, and causal event pairing.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2510.21978

Beyond Reasoning Gains: Mitigating General-Capability Forgetting in Large Reasoning Models

作者:

Hoang Phan ↗Xianjun Yang ↗Yuanshun Yao ↗Jingyu Zhang ↗Shengjie Bi ↗Xiaocheng Tang ↗Madian Khabsa ↗Lijuan Liu ↗Deren Lei ↗

arXiv:2510.21978v2 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has delivered impressive gains in mathematical and multimodal reasoning and has become a standard post-training paradigm for contemporary language and vision-language models. However, the RLVR recipe introduces a significant risk of capability regression, in which models forget foundational skills after prolonged training without employing regularization strategies. We empirically confirm this concern, observing that open-source reasoning models suffer performance degradation on core capabilities such as perception and faithfulness. While imposing regularization terms like KL divergence can help prevent deviation from the base model, these terms are computed on the current task and therefore do not guarantee preservation of broader knowledge. Meanwhile, commonly used experience replay across heterogeneous domains makes it nontrivial to decide how much training emphasis each objective should receive. To address this, we propose RECAP-a replay strategy with dynamic objective reweighting for general knowledge preservation. Our reweighting mechanism adapts online using short-horizon signals of convergence and instability, shifting the post-training focus away from saturated objectives and toward underperforming or volatile ones. Our method is end-to-end and readily applicable to existing RLVR pipelines without training additional models or heavy tuning. Extensive experiments on benchmarks using Qwen2.5-VL-3B and Qwen2.5-VL-7B demonstrate the effectiveness of our method, which not only preserves general capabilities but also improves reasoning by enabling more flexible trade-offs among in-task rewards.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2505.19699

Mosaic: Data-Free Knowledge Distillation via Mixture-of-Experts for Heterogeneous Distributed Environments

作者:

Junming Liu ↗Yanting Gao ↗Yuqi Li ↗Siyuan Meng ↗Yifei Sun ↗Aoqi Wu ↗Yirong Chen ↗Ding Wang ↗Shiping Wen ↗

arXiv:2505.19699v2 Announce Type: replace-cross Abstract: Federated Learning (FL) is a decentralized machine learning paradigm that enables clients to collaboratively train models while preserving data privacy. However, the coexistence of model and data heterogeneity gives rise to inconsistent representations and divergent optimization dynamics across clients, ultimately hindering robust global performance. To transcend these challenges, we propose Mosaic, a novel data-free knowledge distillation framework tailored for heterogeneous distributed environments. Mosaic first trains local generative models to approximate each client's personalized distribution, enabling synthetic data generation that safeguards privacy through strict separation from real data. Subsequently, Mosaic forms a Mixture-of-Experts (MoE) from client models based on their specialized knowledge, and distills it into a global model using the generated data. To further enhance the MoE architecture, Mosaic integrates expert predictions via a lightweight meta model trained on a few representative prototypes. Extensive experiments on standard image and multimodal benchmarks demonstrate that Mosaic consistently outperforms state-of-the-art approaches under both model and data heterogeneity. The source code has been published at https://github.com/Wings-Of-Disaster/Mosaic.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.18394

JetFlow: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting

作者:

Lanxiang Hu ↗Zhaoxiang Feng ↗Yulun Wu ↗Haoran Yuan ↗Yujie Zhao ↗Yu-Yang Qian ↗Bojun Wang ↗Daxin Jiang ↗Yibo Zhu ↗Tajana Rosing ↗Hao Zhang ↗

Speculative decoding (SD) accelerates autoregressive Large Language Models (LLMs) by drafting multiple tokens and verifying them in parallel, but it faces a scaling limitation: increasing the draft budget improves speed only when acceptance remains high and drafting overhead stays low. This ceiling has been difficult to break because prior head-based SD methods face a causality-efficiency dilemma. Autoregressive drafters produce path-conditioned candidates that are effective for tree speculative decoding with higher acceptance length, but their drafting cost grows with tree depth. Bidirectional block-diffusion drafters generate all positions in one pass, but their branch-agnostic marginals can form individually plausible yet mutually inconsistent trees, wasting budget and reducing acceptance. We propose JetFlow, a head-based SD framework that combines one-forward drafting efficiency with branch-wise causal conditioning. JetFlow trains a causal parallel draft head over fused hidden states from the frozen target model, producing candidate trees whose scores align with the target model's autoregressive factorization. This enables JetFlow to convert larger draft budgets into longer accepted prefixes and higher end-to-end speedup. Across math, coding, and chat benchmarks on dense and MoE Qwen3 models, JetFlow consistently outperforms bidirectional-head and tree-based SD baselines. On H100 GPUs, JetFlow achieves up to 9.64x speedup on MATH-500 and 4.58x on open-ended conversational workloads, with further latency gains demonstrated through vLLM integration under realistic serving loads. Our code and models are available at https://github.com/hao-ai-lab/JetFlow.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.CV) 2026-06-12 DOI: arXiv:2606.10683

UniDexTok: A Unified Dexterous Hand Tokenizer from Real Data

作者:

Dong Fang ↗Youjun Wu ↗Yuanxin Zhong ↗Rui Zhang ↗Yunlong Wang ↗Xiaosong Jia ↗Yu-Gang Jiang ↗

Dexterous hands are essential for fine-grained manipulation, but their hardware designs vary substantially across embodiments. Differences in kinematics, joint definitions, and degrees of freedom make it difficult to define a shared state representation compared with parallel grippers. As a result, dexterous-hand data remains fragmented and difficult to use for joint training. In this work, we propose the Unified Dexterous Hand Model (UDHM), which maps human and robot hand states into a shared 22-DoF semantic interface. Based on UDHM, we introduce UniDexTok, a retargeting-free state tokenizer that learns embodiment-conditioned discrete tokens from standardized real joint states. UniDexTok provides a unified representation for heterogeneous dexterous hands without relying on retargeting or simulation data. Compared with the recent baseline UniHM, UniDexTok reduces MPJAE from 15.63 degrees to 0.16 degrees and MPJPE from 18.51 mm to 0.18 mm, corresponding to error reductions of 98.98% and 99.03%, respectively. These results improve reconstruction from centimeter-scale to sub-millimeter accuracy. Experiments further show that data from other embodiments improves target-embodiment reconstruction accuracy, demonstrating the benefit of cross-embodiment tokenization. UniDexTok also shows strong zero-shot and few-shot reconstruction ability when new dexterous hands are introduced.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.08530

GEAR-VLA: Learning Geometry-Aware Action Representations for Generalizable Robotic Manipulation

作者:

Yuan Zhang ↗Shiqi Zhang ↗Yedong Shen ↗Shuai Dong ↗Jiajun Deng ↗Xin Zhang ↗Yuxuan Gao ↗Jiajia Wu ↗Xin Nie ↗Zhiyuan Cheng ↗Jianmin Ji ↗Yanyong Zhang ↗…

arXiv:2606.08530v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models achieve strong benchmark performance but still struggle in real-world deployment with unseen objects, background shifts, and different robot embodiments. We argue that this stems from the lack of a unified geometry-aware manipulation representation, leaving existing VLAs vulnerable to low-level trajectory supervision, misaligned 3D features, and embodiment differences. To address this, we propose GEAR-VLA, a VLA framework for learning unified geometry-aware action representations for generalizable robotic manipulation. GEAR-VLA adopts coarse-to-fine action learning, where multi-source embodied pretraining equips the VLM with embodied reasoning and discrete action understanding before latent action tokens connect action semantics to a gradient-decoupled DiT continuous action expert. It further performs semantic-aligned 3D integration by aligning a trainable 3D spatial backbone with the VLA representation while freezing the original VLM-aligned visual pathway. To share this representation across robots, GEAR-VLA uses embodiment canonicalization, where embodiment-aware states and embodiment-invariant actions confine robot differences to the low-level interface. Extensive simulation and real-world experiments demonstrate strong generalization: GEAR-VLA achieves state-of-the-art performance on LIBERO, zero-shot LIBERO-Plus, and RoboTwin 2.0, reaches 85.9% success on AgileX and 81.0% on the pretraining-unseen LDT-01 embodiment, and obtains 90.1% success on a 6,360-trial universal grasping benchmark with 212 unseen objects. Code and models will be released at https://github.com/babynabeauty/GEAR-VLA.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16101

Effective and Low-cost Lane-based Map Localization for Vehicle-Centric Route Generation

作者:

Hong-Shiang Lin ↗Jung-Hsin Chen ↗Yu-Luen Tzeng ↗Wei-Hao Chen ↗Yi-Chen Lee ↗Li-Jhe Chen ↗Peng-Yuan Chen ↗

Driver-centric route representation plays a vital role in intuitive driving guidance systems. This paper presents OLRA, a low-cost, map-localization-based framework that derives driver-view-aligned routes by matching map-based navigation routes with camera-detected lane markings. This alignment process mutually enhances vehicle localization accuracy and visual route consistency. To bridge the evaluation gap across different paradigms, we introduce practical route evaluation metrics and benchmark OLRA against OpenPilot, a representative direct-generation approach. Experimental results on the nuScenes dataset demonstrate that OLRA outperforms OpenPilot in complex road segments and in route estimation at distance beyond 20 meters, achieving lower overall Euclidean error. This study is expected to promote future research in low-cost, maplocalization-based route generation methods.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17888

MathVis-Fine: Aligning Visual Supervision with Necessity via Progressive Dependency-Guided Training for Multimodal Mathematical Reasoning

作者:

Wanshi Xu ↗Haokun Zhao ↗Haidong Yuan ↗Songjun Cao ↗Long Ma ↗

arXiv:2606.17888v1 Announce Type: new Abstract: Chain-of-Thought (CoT) reasoning has extended from purely linguistic domains to multimodal scenarios; however, existing approaches often treat visual inputs as homogeneous or auxiliary signals, failing to capture the intricate and sample-specific dependencies between text and images in mathematical problem-solving. This gives rise to two core issues: first, the supervisory signals for visual content are generalized and coarse-grained, lacking adaptation to the actual necessity of visual information in each sample; second, training feedback becomes inaccurate when visual rewards are uniformly applied without distinguishing the complementary relationships among inputs. These limitations hinder models from achieving precise multimodal reasoning. In this work, we propose a framework for modeling fine-grained visual dependencies in mathematical reasoning. We first construct the MathVis-Fine dataset, augmenting fine-grained visual annotations with visual dependency ratings. Building upon this dataset, we introduce a two-stage progressive visual enhancement training paradigm that balances answer correctness rewards and visual grounding rewards according to the intrinsic visual dependency level of each sample, thereby mitigating reward bias and improving supervision accuracy. Extensive experiments demonstrate that the MathVis-Fine framework effectively enhances visual perception progressively based on visual dependency, offering a more precise training framework for multimodal mathematical reasoning. We will release the dataset upon acceptance.

阅读与讨论 → 访问原文 →

探索全球前沿学术脉络