MolmoAct2: Action Reasoning Models for Real-world Deployment
Allen Institute for AI（艾伦人工智能研究所）；University of Washington（华盛顿大学）；National University of Singapore（新加坡国立大学）；University of Pennsylvania（宾夕法尼亚大学）；Johns Hopkins University（约翰霍普金斯大学）；Amazon（亚马逊公司）；University of Michigan（密歇根大学）；University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）
Vision-Language-Action (VLA) models aim to provide a single generalist controller for robots, but today's systems fall short on the criteria that matter for real-world deployment. Frontier models are closed, open-weight alternatives are tied to expensive hardware, reasoning-augmented policies pay prohibitive latency for their grounding, and fine-tuned success rates remain below the threshold for dependable use. We present MolmoAct2, a fully open action reasoning model built for practical deployment, advancing its predecessor along five axes. We introduce MolmoER, a VLM backbone specialized for spatial and embodied reasoning, trained on a 3.3M-sample corpus with a specialize-then-rehearse recipe. We release three new datasets spanning low-to-medium cost platforms, including MolmoAct2-BimanualYAM, 720 hours of teleoperated bimanual trajectories that constitute the largest open bimanual dataset to date, together with quality-filtered Franka (DROID) and SO100/101 subsets. We provide OpenFAST, an open-weight, open-data action tokenizer trained on millions of trajectories across five embodiments. We redesign the architecture to graft a flow-matching continuous-action expert onto a discrete-token VLM via per-layer KV-cache conditioning. Finally, we propose MolmoThink, an adaptive-depth reasoning variant that re-predicts depth tokens only for scene regions that change between timesteps, retaining geometric grounding at a fraction of prior latency. In the most extensive empirical study of any open VLA to date, spanning 7 simulation and real-world benchmarks, MolmoAct2 outperforms strong baselines including Pi-05, while MolmoER surpasses GPT-5 and Gemini Robotics ER-1.5 across 13 embodied-reasoning benchmarks. We release model weights, training code, and complete training data. Project page: https://allenai.org/blog/molmoact2
Latent Reasoning VLA: Latent Thinking and Prediction for Vision-Language-Action Models
Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University（西安交通大学人工智能与机器人研究院）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）；Institute of Automation, University of 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）
One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning
School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳）理工学院）；Shenzhen Research Institute of Big Data (SRIBD)（深圳大数据研究院）
该研究推出真实机器人基准One Word, Different Action，评估具身系统的决策不变性与敏感性，发现现代模型在多任务约束整合为机器人动作时存在明显退化，核心挑战是组合多任务需求为正确决策。
Reasoning-VLA: A Fast and General Vision-Language-Action Reasoning Model for Autonomous Driving
Lanzhou University（兰州大学）；National University of Singapore（新加坡国立大学）；University of Science and Technology of China（中国科学技术大学）；Tsinghua University（清华大学）；University of New South Wales（新南威尔士大学）
DualVLA: Building a Generalizable Embodied Agent via Partial Decoupling of Reasoning and Action
MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition, University of Science and Technology of China（脑启发智能感知与认知国家重点实验室，中国科学技术大学）；State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University（多媒体信息处理国家重点实验室，北京大学计算机学院）；CUHK（香港大学）
Do Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoning
Stanford University（斯坦福大学）；Politecnico di Milano（米兰理工大学）；KTH Royal Institute of Technology（皇家理工学院）；Italian Institute of Artificial Intelligence (AI4I)（意大利人工智能研究所）；NVIDIA Research（英伟达研究院）
3DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-training
Shanghai Jiao Tong University（上海交通大学）；Harbin Institute of Technology（哈尔滨工业大学）；Nanyang Technological University（南洋理工大学）；Fudan University（复旦大学）；Nanjing University（南京大学）；Daimon Robotics（达梦机器人）；Great Bay University（大亚大学）
Aux-Think: Exploring Reasoning Strategies for Data-Efficient Vision-Language Navigation
Renmin University of China（中国人民大学）；Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing（北京未来区块链与隐私计算先进创新中心）；Horizon Robotics（地平线机器人）
GRAVA: Grounded Reasoning-to-Action Representation and Learning for Autonomous Driving
National Engineering Research Center of Electric Vehicles, Beijing Institute of Technology（北京理工大学电动车辆国家工程研究中心）；Shenzhen Automotive Research Institute, Beijing Institute of Technology（北京理工大学深圳汽车研究院）；Shenzhen Jiguangzhijie Technology Co., Ltd.（深圳极光智界科技有限公司）；Nanyang Technological University（南洋理工大学）
Reasoning models do not yet follow their reasoning in autonomous driving: The KITScenes LongTail Dataset
Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院）；FZI Research Center for Information Technology（FZI信息技术研究中心）；University Charles III of Madrid（马德里卡洛斯三世大学）；Technical University of Madrid（马德里理工大学）；University of Toronto（多伦多大学）；Delft University of Technology（代尔夫特理工大学）
Cognitive Dual-Process Planning for Autonomous Driving with Structured Scene Knowledge and Verifiable Reasoning-Action Consistency
School of Mechanical Engineering, Beijing Institute of Technology（北京理工大学机械工程学院）；National Engineering Research Center of Electric Vehicles, Beijing Institute of Technology（北京理工大学电动车辆国家工程研究中心）；School of Mechanical Engineering, Southeast University（东南大学机械工程学院）
PreAct-Nav: Agentic Reasoning Before Action for Urban Navigation
Shanghai Jiao Tong University（上海交通大学）；Beihang University（北京航空航天大学）；The Hong Kong University of Science and Technology（香港科技大学）；Tsinghua University（清华大学）；University of Cambridge（剑桥大学）；Intelligent Game and Decision Lab (IGDL)（智能博弈与决策实验室）
Humanoid Agent via Embodied Chain-of-Action Reasoning with Multimodal Foundation Models for Zero-Shot Loco-Manipulation
Embodied AI and Robotics (AIR) Lab, New York University, New York, USA and NYUAD Center for Artificial Intelligence and Robotics, New York University Abu Dhabi, Abu Dhabi, UAE（纽约大学Embodied AI和机器人实验室及纽约大学阿布扎赫尔人工智能与机器人中心）；Harvard AI and Robotics Lab, Harvard University, Boston, USA（哈佛大学人工智能与机器人实验室）；Department of Computer Science, University College London, London, UK（伦敦大学学院计算机科学系）；Department of Computer Science, University of Liverpool, UK（利物浦大学计算机科学系）；NYUAD Center for Artificial Intelligence and Robotics, New York University Abu Dhabi, Abu Dhabi, UAE（纽约大学阿布扎赫尔人工智能与机器人中心）
