Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced the reasoning capabilities of Multimodal Large Language Models (MLLMs), yet how visual evidence is integrated during reasoning remains poorly understood. We explore multimodal RLVR through the lens of cross-modal attention connectivity and find that only a small fraction of tokens (approximately 15%) exhibit strong visual-textual coupling. These high-connectivity tokens act as anchors that ground reasoning in the image, while the majority follow linguistic patterns. During RLVR training, credit assignment naturally concentrates on these anchors, sharpening their visual grounding over time. Building on this insight, we propose Anchor-Token Reinforcement Learning (AT-RL), a lightweight framework that selectively reinforces high-connectivity tokens via graph-based clustering of attention topology. Evaluated across the series (3B-32B), AT-RL introduces only 1.2% overhead yet enables the 32B model to surpass the 72B-Instruct baseline on MathVista (80.2), with consistent gains observed across STEM, video and general tasks. Conversely, training solely on low-connectivity tokens causes severe degradation, confirming that effective multimodal RL hinges on precise credit assignment to visual anchors. Our work reveals that reasoning quality is governed not by token quantity but by the fidelity of cross-modal anchoring.
HKUST(GZ)(香港科技大学(广州));CSE, HKUST(香港科技大学计算机科学与工程系);Xi’an Jiaotong University(西安交通大学);University of Pisa, IT(比萨大学);University of Trento, IT(特伦特大学);Nagoya University(名古屋大学);China University of Mining & Technology, Beijing(中国矿业大学(北京));Tongji University(同济大学);SPIC Energy Science and Technology Research Institute(SPIC能源科学与技术研究院);Shanghai Jiao Tong University(上海交通大学);Fudan University(复旦大学);College of Computing & Data Science, Nanyang Technological University(南洋理工大学计算机与数据科学学院)
When Modalities Conflict: How Unimodal Reasoning Uncertainty Governs Preference Dynamics in MLLMs
Peking University(北京大学);Provable Responsible AI and Data Analytics (PRADA) Lab(可证明负责任的人工智能和数据分析实验室);King Abdullah University of Science and Technology(卡布斯大学);South China University of Technology(华南理工大学);Tsinghua University(清华大学);University of Georgia(佐治亚大学);MBZUAI(马克斯·普朗克人工智能研究所)
Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs
School of Information Science and Engineering, Yunnan University(云南大学信息科学与工程学院);School of Software, Yunnan University(云南大学软件学院);National University of Singapore(新加坡国立大学);School of Engineering, Yunnan University(云南大学工程学院)
Robust Multimodal Large Language Models Against Modality Conflict
School of Artificial Intelligence(人工智能学院);Data Science, University of Science(数据科学,科学大学);Department of Electronic Engineering(电子工程系);Information Science, University of Science(信息科学,科学大学);Huawei Technologies Co., Ltd.(华为技术有限公司)
Merge then Realign: Simple and Effective Modality-Incremental Continual Learning for Multimodal LLMs
Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳));Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies(广东省新型安全智能技术重点实验室)
MODE: Modality-Decomposed Expert-Level Mixed-Precision Quantization for MoE Multimodal LLMs
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Zhongguancun Academy(中关村学院)
OPD-V: Visual On-Policy Self-Distillation with Modality Balance
National University of Singapore(新加坡国立大学);Ludwig Maximilian University of Munich(慕尼黑大学);Munich Center for Machine Learning(慕尼黑机器学习中心);Sun Yat-sen University(中山大学)
Harbin Institute of Technology, Shenzhen, China.(哈尔滨工业大学(深圳));Peng Cheng Laboratory, China.(鹏城实验室);Huazhong University of Science and Technology, China(华中科技大学)
Single-to-mix Modality Alignment with Multimodal Large Language Model for Document Image Machine Translation
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);Zhongke Fanyu Technology Co., Ltd(中科泛语科技有限公司)
SpikeMLLM: Spike-based Multimodal Large Language Models via Modality-Specific Temporal Scales and Temporal Compression
1 Institute of Automation, Chinese Academy of Sciences 2 University of Chinese Academy of Sciences 3 Beijing Academy of Artificial Intelligence 4 Zhongguancun Academy 5 Peking University 6 Key Laboratory of Brain Cognition;Brain-inspired Intelligence Technology 7 Spiking Intelligence Lab, Tianqiao \& Chrissy Chen Institute [0.5em] Equal contribution Corresponding authors
Enhancing Instruction-Following Capability of Visual-Language Models by Reducing Image Redundancy
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);Kuaishou Technology(快手科技);CAIR, HKISI, Chinese Academy of Sciences(中国科学院香港创新研究院人工智能与机器人创新中心)
MACS: Modality-Aware Capacity Scaling for Efficient Multimodal MoE Inference
School of Software, Tsinghua University, Beijing, China(清华大学软件学院,北京,中国);TJUNLP Lab, School of Computer Science and Technology, Tianjin University, China(天津大学计算机科学与技术学院,中国);School of New Media and Communication, Tianjin University, China(天津大学新媒体与传播学院,中国)
LLMs Can Evolve Continually on Modality for X-Modal Reasoning
Dalian University of Technology(大连理工大学);Huawei Noah’s Ark Lab(华为诺亚方舟实验室);Tsinghua University(清华大学);The Hong Kong University of Science and Technology(香港科技大学)
Jailbreaking Multimodal Large Language Models using Multi-Clip Video
Department of Applied Artificial Intelligence, Sungkyunkwan University(应用人工智能系,成均馆大学);Department of Human-Artificial Intelligence Interaction, Sungkyunkwan University(人机交互系,成均馆大学)
A Comprehensive Review of Multimodal Large Language Models: Performance and Challenges Across Different Tasks
School of Computer Science, Northwestern Polytechnical University(西北工业大学计算机学院);School of Automation, Northwestern Polytechnical University(西北工业大学自动化学院);Institute of Medical Research, Northwestern Polytechnical University(西北工业大学医学研究院);School of Computing, The University of Georgia(佐治亚大学计算学院);School of Physics and Information Technology, Shaanxi Normal University(陕西师范大学物理与信息技术学院)
Heuristic-Induced Multimodal Risk Distribution Jailbreak Attack for Multimodal Large Language Models
Sun Yat-Sen University(中山大学);Zhejiang University(浙江大学);BraneMatrix AI;Nanyang Technological University(南洋理工大学);Alibaba Group(阿里巴巴集团);Renmin University of China(中国人民大学)
How to Bridge the Gap between Modalities: Survey on Multimodal Large Language Model
College of Computer, National University of Defense Technology(国防科技大学计算机学院);School of Computer and Information Engineering, Hefei University of Technology(合肥工业大学计算机与信息工程学院)
From Consistency to Complementarity: Aligned and Disentangled Multi-modal Learning for Time Series Understanding and Reasoning
The Hong Kong University of Science(香港科学与技术大学);Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学)
Harnessing Multimodal Large Language Models for Multimodal Sequential Recommendation
Rutgers University(罗格斯大学);University of Science and Technology of China(中国科学技术大学);University of Pennsylvania(宾夕法尼亚大学);Bytedance Inc.(字节跳动公司);Georgia Institute of Technology(佐治亚理工学院);University of Pittsburgh(匹兹堡大学);The Hong Kong University of Science and Technology(香港科技大学)
Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning
The University of British Columbia(英属哥伦比亚大学);Fudan University(复旦大学);Royal College of Science, Imperial College London(伦敦帝国理工学院皇家科学学院);Dyson School of Design Engineering(戴森设计工程学院);Suzhou Institute of Biomedical Engineering and Technology (SIBET), Chinese Academy of Sciences(中国科学院苏州生物医学工程技术研究所);East China Normal University(华东师范大学)
Evolving Safety Landscape of Multi-modal Large Language Models: A Survey of Emerging Threats and Safeguards
University of Alabama at Birmingham(阿拉巴马大学伯明翰分校);NVIDIA(英伟达公司);Penn State University(宾夕法尼亚州立大学);Columbia University(哥伦比亚大学);University of Missouri-Kansas City(密苏里大学堪萨斯分校);Florida State University(佛罗里达州立大学);Auburn University(奥本大学)