Vision-Language Models often struggle with complex visual reasoning due to the visual information loss in textual CoT. Existing methods either add the cost of tool calls or rely on localized patch-based embeddings that are insufficient to extract semantics in multi-step reasoning. We propose "Decompose, Look, and Reason" (DLR), a reinforced latent reasoning framework that dynamically decomposes queries into textual premises, extracts premise-conditioned continuous visual latents, and deduces answers through grounded rationales. We introduce a three-stage training pipeline and propose a novel Spherical Gaussian Latent Policy, to enable effective exploration in the latent space. Extensive experiments on vision-centric benchmarks show that DLR consistently outperforms strong baselines, including text-only, interleaved multimodal CoT, and latent reasoning methods, while providing superior stepwise interpretability.
Multi-Turn Multi-Agent Dialogue for Collaborative Reconstruction Improves VLM Performance on Spatial Reasoning, But Only Barely
Computational Linguistics, Department of Linguistics University of Potsdam(语言学计算系,语言学系 柏林洪堡大学);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心(DFKI))
When Words Speak Louder than Images: Towards Understanding Language Bias in Vision-Language Models
University of Waterloo(滑铁卢大学);Johns Hopkins University(约翰霍普金斯大学);University of California, Santa Barbara(加州大学圣塔芭芭拉分校);Nanyang Technological University(南洋理工大学);Indiana University(印第安纳大学)
Probing and Inducing Combinational Creativity in Vision-Language Models
Institute for Artificial Intelligence, Peking University(北京大学人工智能研究院);Yuanpei College, Peking University(北京大学元培学院);State Key Laboratory of General Artificial Intelligence, BIGAI(人工智能通用基础理论国家重点实验室);Center on Frontiers of Computing Studies, School of Computer Science, Peking University(计算机科学学院前沿计算研究中心)
Enhancing Abnormality Grounding for Vision Language Models with Knowledge Descriptions
Technical University of Munich(慕尼黑工业大学);Munich Center for Machine Learning(慕尼黑机器学习中心);Helmholtz AI(亥姆霍兹人工智能研究所);Helmholtz Munich(亥姆霍兹慕尼黑中心);Imperial College London(伦敦帝国学院);King’s College London(伦敦国王学院)
SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes
The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州));Zhongguancun Academy(中关村学院);Tsinghua University(清华大学);Helsinki University(赫尔辛基大学)
Bridging the 2D-3D Gap: A Hierarchical Semantic-Geometric Map for Vision Language Navigation
School of Computer Science and Technology, East China Normal University(东华大学计算机科学与技术学院);Bosch Corporate Research(博世企业研究);King Abdullah University of Science and Technology(卡布斯大学)
AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play
Emory University(埃默里大学);Georgia Institute of Technology(佐治亚理工学院);Rutgers University(罗格斯大学);SUNY Albany(纽约州立大学阿尔巴尼分校);UT Southwestern Medical Center(德克萨斯西南医学中心)