Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning
University of Cambridge(剑桥大学);Pioneer Center for AI, University of Copenhagen(人工智能先锋中心,哥本哈根大学);Hong Kong University of Science(香港科学大学);University of California San Diego(加州大学圣地亚哥分校);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Peking University(北京大学)
Chengzu Li, Zanyi Wang, Jiaang Li, Yi Xu, Han Zhou, Huanyu Zhang, Ruichuan An, Dengyang Jiang, Zhaochong An, Ivan Vulić, Serge Belongie, Anna Korhonen
备注:
8 pages, 3 figures, 3 tables (26 pages, 13 figures, 6 tables including references and appendices)
Vision-Language Models have excelled at textual reasoning, but they often struggle with fine-grained spatial understanding and continuous action planning, failing to simulate the dynamics required for complex visual reasoning. In this work, we formulate visual reasoning by means of video generation models, positing that generated frames can act as intermediate reasoning steps between initial states and solutions. We evaluate their capacity in two distinct regimes: Maze Navigation for sequential discrete planning with low visual change and Tangram Puzzle for continuous manipulation with high visual change. Our experiments reveal three critical insights: (1) Robust Zero-Shot Generalization: In both tasks, the model demonstrates strong performance on unseen data distributions without specific finetuning. (2) Visual Context: The model effectively uses visual context as explicit control, such as agent icons and tangram shapes, enabling it to maintain high visual consistency and adapt its planning capability robustly to unseen patterns. (3) Visual Test-Time Scaling: We observe a test-time scaling law in sequential planning; increasing the generated video length (visual inference budget) empowers better zero-shot generalization to spatially and temporally complex paths. These findings suggest that video generation is not merely a media tool, but a scalable, generalizable paradigm for visual reasoning.
VisuoThink: Empowering LVLM Reasoning with Multimodal Tree Search
Fudan University(复旦大学);Shanghai Innovation Institute(上海创新研究院);The University of Sydney(悉尼大学);Shanghai AI Laboratory(上海人工智能实验室);Nanyang Technological University(南洋理工大学)
Don't Wait to Reply: Towards Responsive yet Thoughtful Dialogue through Proactive Thinking
Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua University(清华大学计算机科学与技术系、人工智能研究院);Institute for AI Industry Research (AIR), Tsinghua University(清华大学人工智能产业研究院);Tencent(腾讯)
Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);University of Chinese Academy of Sciences(中国科学院大学);Ant Group(蚂蚁集团);CUHK MMLab(香港中文大学MMLab);Nanjing University(南京大学)
Thinking with Imagination: Agentic Visual Spatial Reasoning with World Simulators
The University of Hong Kong(香港大学);Shanghai AI Laboratory(上海人工智能实验室);Shanghai Jiao Tong University(上海交通大学);Fudan University(复旦大学);Beihang University(北航大学)
Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning
Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China(中国人民大学人工智能学院);Beijing Key Laboratory of Research on Large Models(北京大型模型研究关键实验室);Engineering Research Center of Next-Generation Intelligent Search(下一代智能搜索工程研究中心);College of Artificial Intelligence, Tianjin University of Science(天津科技大学人工智能学院)
LogReasoner: Empowering LLMs with Expert-like Coarse-to-Fine Reasoning for Automated Log Analysis
Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University(上海数据科学关键实验室,计算机科学学院,复旦大学);School of Artificial Intelligence and Computer Science, Jiangnan University(人工智能与计算机科学学院,江南大学)
Aligning Thoughts with Answers: Probability Rewards to Tame Thinking Drift
National University of Singapore(新加坡国立大学);A*STAR Institute of Advanced Intelligence and Computing, Singapore(新加坡科技研究局先进智能与计算研究所);South China University of Technology(华南理工大学);University of Science and Technology of China(中国科学技术大学);Google DeepMind(谷歌DeepMind)
How and What to Imagine? Visual Thinking in Unified Multimodal Models for Cross-View Spatial Reasoning
Mila - Québec AI Institute(蒙特利尔AI研究所);Université de Montréal(蒙特利尔大学);McGill University(麦吉尔大学);ServiceNow AI Research(ServiceNow人工智能研究);Canada CIFAR AI Chair(加拿大CIFAR人工智能主席)
LLM-hRIC: LLM-empowered Hierarchical RAN Intelligent Control for O-RAN
School of Electrical and Electronic Engineering, Yonsei University(延世大学电气与电子工程学院);Information System Technology and Design, Singapore University of Technology and Design(新加坡科技与设计大学信息系统技术与设计)
Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Large Language Model Department, Tencent(腾讯大语言模型部);University of Electronic Science and Technology of China(电子科技大学);Hong Kong University of Science and Technology(香港科技大学);Zhongguancun Academy(中关村学院)