Latent action learning infers pseudo-action labels from visual transitions, providing an approach to leverage internet-scale video for embodied AI. However, most methods learn latent actions without structural priors that encode the additive, compositional structure of physical motion. As a result, latents often entangle irrelevant scene details or information about future observations with true state changes and miscalibrate motion magnitude. We introduce Additively Compositional Latent Action Model (AC-LAM), which enforces scene-wise additive composition structure over short horizons on the latent action space. These AC constraints encourage simple algebraic structure in the latent action space~(identity, inverse, cycle consistency) and suppress information that does not compose additively. Empirically, AC-LAM learns more structured, motion-specific, and displacement-calibrated latent actions and provides stronger supervision for downstream policy learning, outperforming state-of-the-art LAMs across simulated and real-world tabletop tasks.
RoboFactory: Exploring Embodied Agent Collaboration with Compositional Constraints
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Oxford（牛津大学）；HKU（香港大学）；Sun Yat-sen University（中山大学）；Tongji University（同济大学）；Shanghai Jiao Tong University（上海交通大学）
SkillMemo: Expert-guided Skill Memory Framework for Compositional Embodied Manipulation
Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Department of Automation, Tsinghua University（清华大学自动化系）；Nanyang Technological University（南洋理工大学）；Beijing Normal University（北京师范大学）
CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environment
Shanghai Jiao Tong University（上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；Harbin Institute of Technology（哈尔滨工业大学）；University of Science and Technology of China（中国科学技术大学）；CUHK-Shenzhen（香港中文大学（深圳））；Fudan University（复旦大学）；Dalian University of Technology（大连理工大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Oxford（牛津大学）
Sparse Compositional Flow Matching by geometric assembly from motion primitives
Tsinghua Shenzhen Graduate School, Tsinghua University, Shenzhen, China（清华大学深圳研究生院，清华大学，深圳，中国）；School of AI, Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳）人工智能学院）
University of California, Santa Cruz（加州大学圣克鲁兹分校）；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）
Benefiting from language flexibility and compositionality, humans naturally intend to use language to command an embodied agent for complex tasks such as navigation and object manipulation. In this work, we aim to fill the blank of the last mile of embodied agents -- object manipulation by following human guidance, e.g., "move the red mug next to the box while keeping it upright." To this end, we introduce an Automatic Manipulation Solver (AMSolver) system and build a Vision-and-Language Manipulation benchmark (VLMbench) based on it, containing various language instructions on categorized robo…
EmbodiedCLBench: Evaluating Continual Learning for Self-Evolving Embodied Agents
School of Intelligence Science and Technology, Peking University（北京大学智能科学与技术学院）；State Key Laboratory of General Artificial Intelligence, BIGAI（北京通用人工智能研究院通用人工智能国家重点实验室）；University of Hong Kong（香港大学）
Socratic Planner: Self-QA-Based Zero-Shot Planning for Embodied Instruction Following
Seoul National University（首尔大学）；Interdisciplinary Program in AI (IPAI), Seoul National University（首尔大学人工智能交叉项目）；AI Institute, Seoul National University（首尔大学人工智能研究院）
Learning Primitive Embodied World Models: Towards Scalable Robotic Learning
Shanghai AI Lab（上海人工智能实验室）；Fudan（复旦大学）；SJTU（上海交通大学）；NJUST（南京理工大学）；THU（清华大学）；Harvard（哈佛大学）；ZJU（浙江大学）；NJU（南京大学）；USTC（中国科学技术大学）；Tongji（同济大学）；HKUST (GZ)（香港科技大学（广州））
COMBO: Compositional World Models for Embodied Multi-Agent Cooperation
University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）；IIIS, Tsinghua University（清华大学信息科学技术学院）；Peking University（北京大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；Honda Research Institute USA（美国本田研究所）；MIT（麻省理工学院）
FluxVLA Engine: A One-Stop VLA Engineering Platform for Embodied Intelligence
LimX Dynamics；Nankai University（南开大学）；The University of Hong Kong（香港大学）；University of Electronic Science and Technology of China（电子科技大学）；Beijing Institute of Technology（北京理工大学）；Xidian University（西安电子科技大学）；Zhejiang University（浙江大学）；Xi’an Jiaotong University（西安交通大学）
Compromising Embodied Agents with Contextual Backdoor Attacks
Beihang University（北京航空航天大学）；National University of Singapore（新加坡国立大学）；Zhongguancun Laboratory（中关村实验室）；Nanyang Technological University（南洋理工大学）；University of Science and Technology of China（中国科学技术大学）；A*STAR（新加坡科技研究局）
VLA-Trace: Diagnosing Vision-Language-Action Models through Representation and Behavior Tracing
University of Science and Technology of China（中国科学技术大学）；University of Manchester（曼彻斯特大学）；Beihang University（北航）；Fudan University（复旦大学）；University of New South Wales（新南威尔士大学）
SleepWalk: A Three-Tier Benchmark for Stress-Testing Instruction-Guided Vision-Language Navigation
Indian AI Research Organization (IAIRO)（印度人工智能研究组织）；ĸragya Lab, BITS Pilani Goa（BITS Pilani Goa 的 ĸragya 实验室）；University of Dhaka（达卡大学）；Delhi Technological University（德里技术大学）；Apple（苹果公司）；Meta
Sun Yat-sen University（中山大学）；Shenzhen Loop Area Institute（深圳河套学院）；Baidu Inc.（百度公司）；Cardiff University（卡迪夫大学）；Guangdong Key Laboratory of Big Data Analysis and Processing（广东大数据分析与处理重点实验室）
