Reinforcement learning is a promising approach for improving the capabilities of vision-language-action (VLA) models while avoiding the heavy data requirements of imitation learning. However, its effectiveness for VLA models is often constrained by sparse supervision and the difficulty of designing informative reward signals for long-horizon manipulation. In this work, we present Feat2Go, a fine-grained value estimation framework for embodied reinforcement learning. Specifically, Feat2Go first derives a continuous progress target from a pretrained visual world model by measuring patch-level similarity to subgoal states and partitioning episodes into semantic stages with trend-based clustering. We then train an embodied value model to predict this structural progress from the current observation and task instruction, and use the predicted value to reshape terminal rewards during policy optimization. The proposed framework is compatible with existing VLA policy reinforcement learning pipelines, including PPO and GRPO, and does not rely on manual reward engineering. Extensive experiments on ManiSkill3 and RoboTwin 2.0 demonstrate that Feat2Go consistently improves the performance of existing VLA models under both single-arm and bimanual manipulation settings. More specifically, on ManiSkill3, Feat2Go improves OpenVLAOFT from 17.5% to 82.9% average out-of-distribution success while retaining 96.9% in-distribution performance. On RoboTwin 2.0, Feat2Go achieves an average success rate of 88.8% in domain-randomized task settings, outperforming prior reinforcement learning methods.
PRICE the Action Chunks: Physical Relational Credit Assignment for Embodied Reinforcement Learning
Fudan University(复旦大学);Shanghai Innovation Institute(上海创新研究院);Beihang University(北京航空航天大学);Simple AI;Tuojing Intelligence(拓境智能);Tsinghua University(清华大学);The University of Hong Kong(香港大学)
3D Generation for Embodied AI and Robotic Simulation: A Survey
Hong Kong University of Science and Technology(香港科技大学);Wuhan University(武汉大学);Harbin Institute of Technology(哈尔滨工业大学);Xinjiang University(新疆大学);Tencent(腾讯)
VIKI-R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning
Shanghai Jiao Tong University(上海交通大学);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);University of Science and Technology of China(中国科学技术大学);University of Oxford(牛津大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))
ERQA-Plus: A Diagnostic Benchmark for Reasoning in Embodied AI
Centre for Frontier AI Research, Agency for Science, Technology and Research(新加坡科技研究局前沿人工智能研究中心);College of Computing and Data Science, Nanyang Technological University(南洋理工大学计算与数据科学学院)
GuardianBench: A Same-Scene Instruction-Contrastive Benchmark for Latent Contextual Risk in Embodied AI
Shandong University(山东大学);National University of Singapore(新加坡国立大学);Nanjing University of Aeronautics and Astronautics(南京航空航天大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Xiaomi Corporation(小米公司)
NEXUS: Learning Safety Constraints and Action Models for Safe and Robust Embodied Planning
School of Artificial Intelligence and Automation(人工智能与自动化学院);Huazhong University of Science and Technology(华中科技大学);School of Software(软件学院);Henan University(河南大学);Institute of Information Engineering(信息工程研究所);Chinese Academy of Sciences(中国科学院);School of Cyberspace Security(网络空间安全学院);University of the Chinese Academy of Sciences(中国科学院大学);Peking University(北京大学)
EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI
The Chinese University of Hong Kong(香港中文大学);Tsinghua University(清华大学);The University of Hong Kong(香港大学);Shanghai Jiao Tong University(上海交通大学);Shanghai AI Laboratory(上海人工智能实验室)
Plan in Sandbox, Navigate in Open Worlds: Learning Physics-Grounded Abstracted Experience for Embodied Navigation
School of Computing and Artificial Intelligence, Southwest Jiaotong University, China(计算机与人工智能学院,西南交通大学,中国);Centre for Frontier AI Research A*STAR, Singapore(前沿人工智能研究A*STAR中心,新加坡);School of Computer Science, University of Leeds, UK(计算机科学学院,利兹大学,英国)
ReactHuman: A Physics-Grounded Benchmark for Human-Like Reactive Decision-Making in Embodied Multimodal LLMs
Université de Montréal(蒙特利尔大学);Mila – Quebec AI Institute(米拉-魁北克人工智能研究所);McGill University(麦吉尔大学);McMaster University(麦克马斯特大学);Meta Platforms(Meta平台公司)
LaGEA: Language Guided Embodied Agents for Robotic Manipulation
Department of Robotics;Mechatronics Engineering, University of Dhaka, Bangladesh;Center for Computational \& Data Sciences, Independent University, Bangladesh
Information Theoretically Aided Reinforcement Learning for Embodied Agents
Max Planck Institute for Mathematics in the Sciences(马克斯·普朗克数学科学研究所);Faculty of Mathematics and Computer Science, Leipzig University(莱比锡大学数学与计算机科学学院);Santa Fe Institute(圣塔菲研究所)
SVLL: Staged Vision-Language Learning for Physically Grounded Embodied Task Planning
FNii-Shenzhen, CUHKSZ(FNii深圳,CUHKSZ);SSE, CUHKSZ(SSE,CUHKSZ);SAI, CUHKSZ(SAI,CUHKSZ);Shenzhen University(深圳大学);University of Sydney(悉尼大学);Harbin Engineering University(哈尔滨工程大学);Ising AI
UniPart: Towards Zero-shot Language-Grounded 3D Part Segmentation for Embodied Interaction
NLPR, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所模式识别国家重点实验室);Peking University(北京大学);Tsinghua University(清华大学);Galbot Inc.(北京银河通用机器人有限公司);Shanghai Jiao Tong University(上海交通大学);Beijing Academy of Artificial Intelligence(北京智源人工智能研究院)