Overview

Latent action learning studies how to represent meaningful behavior compactly, often by inferring actions from video or trajectories that lack action labels. Common approaches use inverse and forward dynamics, predictive world models, vector-quantized or continuous flow-based representations, and geometric or hierarchical structure. These representations can help vision-language-action policies plan and control robots, especially across long tasks, different embodiments, and limited demonstration data; LIBERO, CALVIN, and RoboTwin are common evaluation settings alongside real robots. A central open tension is that representations good at reconstructing visual changes may capture nuisance or environmental variation rather than controllable actions, so predictive consistency and proxy metrics do not always guarantee better downstream control.

Latest papers

CardsList

123 papers15 papers in the last four weeks, level with the four weeks before. 0.1% of all new papers.

  1. Beyond Prediction Gain: LASR as a Metric for Action-Grounded Latents in World Models

    Nov 13, 2026Yu Zhou, Changsong Lei, Yixiang Wei +6World Model EvaluationLatent Action Learning

  2. SALA: Semantically Aligned Latent Actions Enable Dual-Axis Scaling of Embodied Foundation Models

    Nov 12, 2026Jiachun Jin, Zetong Zhou, Haonan Wang +1Robot Learning from VideoLatent Action Learning

  3. Latent RL Priors for Data-Efficient Dexterous Manipulation

    Nov 5, 2026Yixiao Wang, Peiyi Chen, Mustafa Mukadam +3Robotic ManipulationData-Efficient Robot Learning

  4. ELA: End-to-End Latent Action Policy Learning from Video Demonstrations

    Nov 3, 2026Cheng-Chun Hsu, Peter StoneLatent Action LearningRobot Imitation Learning

  5. SurgTransfer: Multi-Step Latent Action Modeling for Action Transfer in Surgical World Model

    Nov 3, 2026Saurabh Koju, Love Panta, Deepak Thapa +6Latent Action LearningWorld Model Learning

  6. OpenNWM: A Generalist Navigation World Model with Latent Action Pretraining

    Nov 1, 2026Dujun Nie, Ruiqi Song, Zixuan Wang +6World Model LearningRobot Navigation

  7. RynnVLA-Latent: Scaling Robotic Manipulation Foundation Models with 3D-Grounded Latent Actions

    Oct 25, 2026Haoyu Zhao, Xingyue Zhao, Zhian Su +11VLA Model TrainingRobot Foundation Models

  8. EXE: Egocentric Cross-Embodiment Pretraining via Factorized Latent Actions for VLA Models

    Oct 24, 2026Qingyuan Cai, Bo Zhang, Yongzhen Huang +1Cross-Embodiment Robot LearningRobot Learning from Video

  9. Skeleton-Grounded Latent Action Model for Robot World Models

    Oct 22, 2026Qi Lv, Jiayi Lin, Yuxin Jiang +3World Models for RoboticsLatent Action Learning

  10. Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models

    Oct 20, 2026Zhiyuan Liu, Sicheng Liu, Haiyu WuQuantum Circuit Born MachinesAction-Conditioned World Models

  11. Multi-View Latent Action Pretraining for Cross-Embodiment Robot Learning from Videos

    Oct 20, 2026Oliver Hausdörfer, Felix Minzenmay, Mani Majd +3Robot Learning from VideoLatent Action Learning

  12. A latent control model for realistic rodent motion

    Oct 19, 2026Aidan Sirbu, Charles Y. Zhang, Yuanjia Yang +3Latent Action LearningAnimal Behavior Analysis

  13. TS-WAM: Temporal Latent Actions and World-Action Coupling for Vision-Language Robot Control

    Oct 15, 2026Menghao Yin, Jiaju Yin, Xiyue Dong +5Vision-Language-Action ModelsWorld Action Models

  14. What Should Robot Policies Learn to Predict? Rethinking Supervision Targets in Vision-Language-Action Learning

    Oct 15, 2026Haitao Shen, Hanyu Liu, Yihan Lin +7Vision-Language-Action ModelsVisuomotor Policy Learning

  15. EgoJEPA: Beyond Retargeting for Dexterous Manipulation from Egocentric Videos

    Oct 14, 2026Meng Cao, Can Zhao, Jinjie Wei +12Robotic ManipulationCross-Embodiment Robot Learning

  16. OneLA: Learning Many Worlds from Latent Actions

    Oct 14, 2026Ziyu Zhu, Zhaoxu Sun, Fan Jiang +3World Model LearningAction-Conditioned World Models

  17. Interaction Latents as Actions for Contact-Rich Manipulation

    Oct 13, 2026Yuhang Zheng, Yujie Zang, Yuchen Liu +9Robot Policy LearningContact-Rich Robotic Manipulation

  18. UniLAP: A Unified Latent Action Policy for Cross-Embodiment Manipulation

    Oct 12, 2026Lei Yao, Qiao Feng, Mutian Tong +3Latent Action LearningRobotic Manipulation

  19. LAPT-Scale: Scaling Embodied Policies with Latent Action Pre-Training on Human Videos

    Oct 9, 2026Zhiyuan Xiang, Peidong Liu, Mingyang Li +4Latent Action LearningData-Efficient Robot Learning

  20. VioLA: Learning Generalist Humanoid Control Policies from Human Data

    Oct 8, 2026Mert Albaba, Jens Beißwenger, Anna Manasyan +6Humanoid Robot ControlRobot Imitation Learning

  21. Cross-Embodiment Robot Foundation World Models with Latent Actions

    Oct 7, 2026Huang Huang, Sriram Yenamandra, Arjun Majumdar +9World Models for RoboticsAction-Conditioned World Models

  22. TERRA: Learning Transportable Latent Actions through Temporal Effect Representation and Relational Alignment

    Oct 7, 2026Tianxingjian Ding, Mubarak Shah, Yu TianLatent Action LearningRobot Learning

  23. Flow Policies as Actions of Skill-Level World Models: Learned and Symbolic Abstractions for Long-Horizon Planning

    Oct 3, 2026Andreu Matoses Gimenez, Andrei-Carlo Papuc, Chris Pek +1World Model-Based PlanningRobot Skill Learning

  24. World Action Learning via Interaction-Centric Spectral Latent Guidance

    Oct 2, 2026Zhiming Liu, Yikun Miao, Ying Chen +7Robot Policy LearningCross-Embodiment Robot Learning