Multimodal in-context learning (ICL) is an important capability of multimodal large language models (MLLMs). Its performance is closely tied to the demonstrations it uses, making demonstration retrieval a core problem for ICL. Common methods use fixed scoring methods to retrieve all the demonstrations at once, which may introduce redundancy and noise. Context-aware retrieval methods have emerged to address this problem by allowing incremental retrieval from a growing context. However, none of them lets the MLLM learn how to judge what new demonstrations it needs and then carry out progressive retrieval on its own. To this end, we introduce ActICL, which trains an MLLM agent to acquire and use an ICL context via tool-augmented reasoning. An ActICL agent analyzes its own information gaps during iterative reasoning and tool-use actions. Based on these gaps, it performs targeted retrieval to acquire a set of initial candidates. It can then use the existing context to further filter the candidates and refine the context, or it can iteratively perform a new retrieval. When the agent judges that the context is sufficient for the task, or when the resource limit is reached with the user request met, it sends the acquired context to an answering model, either the agent itself or another MLLM. We use a two-stage post-training procedure that jointly optimizes the agent's acquisition and answering. With an adaptive number of demonstrations, ActICL improves average performance by 11.6 percentage points over standard similarity-based retrieval across seven benchmarks and three MLLMs, and by 4.4 points compared to the strongest prior baseline. The contexts acquired by ActICL agents can also be used effectively by frontier models. When GPT-5.6-Sol performs 4-shot ICL, ActICL's contexts improve its performance by 3.5 points over standard retrieval, averaged over three benchmarks.
A Survey on Agentic Multimodal Large Language Models
Nanyang Technological University, Singapore(南洋理工大学,新加坡);Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳));Shenzhen Research Institute of Big Data, China(大数据研究 institute);Sun Yat-sen University, China(中山大学);City University of Hong Kong, China(香港城市大学);Communication University of China, China(通信大学)
V-Retrver: Evidence-Driven Agentic Reasoning for Universal Multimodal Retrieval
Tsinghua University(清华大学);University of Central Florida(中央佛罗里达大学);Fudan University(复旦大学);The Australian National University(澳大利亚国立大学);The University of Hong Kong(香港大学);Pengcheng Laboratory(鹏城实验室);Central South University(中南大学)
ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning
Fudan University(复旦大学);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Zhejiang University(浙江大学);Shanghai Jiao Tong University(上海交通大学);The Chinese University of Hong Kong(香港中文大学);Shanghai Innovation Institute(上海创新研究院)
Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization
National Key Laboratory for Novel Software Technology, Nanjing University, China(南京大学计算机软件新技术国家重点实验室);School of Artificial Intelligence, Nanjing University, China(南京大学人工智能学院);AI Business, Alibaba Group(阿里巴巴集团智能计算研究院);School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, China(上海交通大学自动化与智能感知学院);School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen, China(中山大学智能工程学院(深圳));School of Information Science and Technology, University of Science and Technology of China(中国科学技术大学信息科学技术学院);School of Software Technology, Zhejiang University, China(浙江大学软件学院);School of Computing, National University of Singapore, Singapore, Singapore(新加坡国立大学计算机学院)
PathFound: An Agentic Multimodal Model Activating Evidence-seeking Pathological Diagnosis
Qing Yuan Research Institute, Shanghai Jiao Tong University(上海交通大学庆元研究院);Shanghai Innovation Institute(上海创新研究院);Department of Pathology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China(中国科学技术大学生命科学与医学学院病理科);Intelligent Pathology Institute, Division of Life Sciences and Medicine(生命科学与医学学院智能病理研究所);Department of Pathology, Fudan University Shanghai Cancer Center(复旦大学上海癌症中心病理科);Department of Oncology, Shanghai Medical College, Fudan University(复旦大学上海医学院肿瘤科);Institute of Pathology, Fudan University(复旦大学病理研究所);Department of Pathology, The First Affiliated Hospital with Nanjing Medical University(南京医科大学第一附属医院病理科)
VideoGAIA: A Benchmark for General AI Assistants on Agentic Video Understanding
The Chinese University of Hong Kong(香港中文大学);Ant Group(蚂蚁集团);Tsinghua University(清华大学);Tongji University(同济大学);The Hong Kong University of Science and Technology(香港科技大学)
Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning
Arizona State University(亚利桑那州立大学);Clemson University(克莱姆森大学);Washington University in St. Louis(圣路易斯华盛顿大学);Halmstad University(哈姆斯塔德大学);Florida State University(佛罗里达州立大学);Rice University(里士满大学)
EvoGraph-R1: Self-Evolving Multimodal Knowledge Hypergraphs for Agentic Retrieval
Northwestern Polytechnical University(西北工业大学);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);The University of Hong Kong(香港大学);Monash University(莫纳什大学);The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳))
LLM4Cell: A Survey of Large Language and Agentic Models for Single-Cell Biology
Department of Computer Science, Virginia Tech, Blacksburg, VA, USA(弗吉尼亚理工大学计算机科学系);Department of Computational Modeling and Data Analytics, Virginia Tech, Blacksburg, VA, USA(弗吉尼亚理工大学计算建模与数据分析系);Institute of Information and Technology, University of Dhaka, Dhaka, Bangladesh(达卡大学信息技术学院);Fralin Biomedical Research Institute at VTC: Cancer Research Center, Washington DC, USA(弗拉林生物医学研究中心:癌症研究中心)
Octopus: Agentic Multimodal Reasoning with Six-Capability Orchestration
Sun Yat-sen University(中山大学);Shanghai Jiao Tong University(上海交通大学);Zhejiang University(浙江大学);Tsinghua University(清华大学);Peking University(北京大学);The Hong Kong University of Science and Technology(香港科技大学);University of Chinese Academy of Sciences(中国科学院大学)
UrbanWorld2.0: A Multimodal Agentic Framework for Reality-Aligned 3D World Generation at City-Scale
College of AI, Tsinghua University(人工智能学院,清华大学);Shenzhen International Graduate School, Tsinghua University(深圳国际研究生院,清华大学);Department of Electronic Engineering, BNRist, Tsinghua University(电子工程系,北京研究院,清华大学)