Efficient Inference for World Action Models

Overview

World action models predict actions together with how the world may change, and efficient inference studies how to run these models within real-time and memory limits. Approaches include using compact latent, motion, or object representations instead of dense video, distilling models to generate actions in few steps, and reusing cached computations across control cycles. Systems also use pruning, quantization, asynchronous streaming, and adaptive compute, with evaluation on robot manipulation benchmarks such as LIBERO and RoboTwin as well as real robots and navigation tasks. The central tension is that future prediction can aid robustness and long-horizon planning, but generating or reusing it adds cost and can introduce stale or inaccurate information.

Latest papers

CardsList

103 papers26 papers in the last four weeks, down 40% on the four weeks before. 0.2% of all new papers.

  1. Distilling World Action Models for Efficient Bimanual Manipulation via Anchored Path Matching

    Nov 17, 2026Jialong Liu, Junjie He, Yandu Chen +5Diffusion Model DistillationEfficient Inference for World Action Models

  2. Towards Data-Efficient, Real-Time World Models

    Nov 17, 2026Raluca Stevenson, David Bignell, Jennifer C. White +9World Model LearningGame World Models

  3. Manufacturing Differential Sparsity: Planner-Hardware Co-Design for Efficient World Model Planning

    Nov 17, 2026Wenbo Zhu, Paul Yi-Chia Chen, Yuanhan Li +2Hardware-Software Co-DesignWorld Model-Based Planning

  4. AR-WAM: efficient streaming world-action modeling for autonomous driving

    Nov 16, 2026Baiting Luo, Shatadal Mishra, Ahmadreza Moradipari +1World Models for Autonomous DrivingEnd-to-End Autonomous Driving

  5. ResWAM: Learning What Changes, Not What Stays, for World-Action Models

    Nov 14, 2026Chongyu Fan, Gaowen Liu, Ramana Rao Kompella +2World Action ModelsWorld Models for VLA Models

  6. WA-JEPA: Goal-Aware Predictive Feedback Action Refinement for a World-Action Model

    Nov 14, 2026Tao An, Weihang Liang, Tianqi Wang +7Action-Conditioned World ModelsEfficient Inference for World Action Models

  7. Bind-WAM: Binding-Uncertainty-Guided Selective Future Imagination in World Action Models

    Nov 11, 2026Chenhao Qiu, Shuyi Chen, Songen Gu +6World Action ModelsWorld Models for VLA Models

  8. From Video Frames to Robot Meta-Frames: A Unified Latent Interface for World Action Models

    Nov 10, 2026Guoqiang Liang, Yiqi Lin, Yanzhe Chen +2World Action ModelsWorld Models for VLA Models

  9. Elastic-WAM: Adaptive Spatiotemporal Imagination for Efficient Robot Control

    Nov 10, 2026Haitian Li, Tianfu Li, Xuxi Zhao +8World Models for RoboticsRobotic Control

  10. WAM is capable in BEHAVIOR-1K: the 2nd-place solution to the 2026 BEHAVIOR Challenge

    Nov 10, 2026Yelin He, Jiaquan Ye, Xinming Wei +5Mobile ManipulationVLA Model Adaptation

  11. Planners Exploit Deterministic Error: Dithered Low-Bit World Models

    Nov 9, 2026Yong Hwi Song, Keon Myung LeeWorld Model-Based PlanningLow-Bit Quantization

  12. ForeAct: Geometry-Guided Action Learning from World Representations

    Nov 8, 2026Wei Lu, Jiaxi Zhang, Yunheng Wang +8Robotic ControlRobotic Manipulation

  13. La-JEPA: Compact Robot Policies with Latent World Modeling

    Nov 5, 2026Mateo Clémente, George Z. Xue, Amin Abyaneh +5Robot Policy LearningWorld Action Models

  14. HiLo-WAM: Decoupling Long-Horizon Task Evolution and Local Dynamics in World Action Models

    Nov 3, 2026Xingyu Wang, Yanjiang Guo, Zhijing Ding +7World Action ModelsWorld Models for Robotics

  15. PrimeSim: A Generalist World Model Simulator from Large-Scale Cross-Embodiment Demonstrations

    Nov 3, 2026Travis Davies, Wentao Ma, Lihao Guo +10World Models for RoboticsAction-Conditioned World Models

  16. Seeing Through the Illusion: Test-Time Scaling for World Action Models

    Nov 2, 2026Cheikh Ahmed, Mahdi Mostajabdaveh, Milad Yazdani +1World Models for RoboticsEfficient Inference for World Action Models

  17. W2A: Distilling World Priors into Action Models for Real-Time Robotic Control

    Nov 2, 2026Xiaopeng Zhang, Yongjin Mu, Yueyang Weng +2World Action ModelsReal-Time Robot Control

  18. Utility-WAM: Correcting Future-Induced Action Changes

    Oct 29, 2026Feihong Zhang, Xi Lin, Yulong Shi +7World Action ModelsWorld Models for VLA Models

  19. TurboWAM: Learning What Matters for Fast World Action Models

    Oct 24, 2026Guile Wu, Shuangzhi Li, Xingxin Chen +5Efficient Inference for World Action ModelsWorld Action Models

  20. De-Diffusing WAMs: Action-Sufficient Direct Prediction for Robot Control

    Oct 20, 2026Shaoyi Zheng, Zhongtao Jiang, Yue Peng +4Efficient Inference for World Action ModelsWorld Models for VLA Models

  21. LaLa-World: Scaling Latent World Models with Natural Language Actions

    Oct 20, 2026Junha Song, Geonmo Gu, Jeongeun Park +5World Model LearningLatent World Models

  22. TrajWAM: An Efficient Trajectory-Guided World Action Model for Dynamic Robotic Manipulation

    Oct 19, 2026Haohan Geng, Sha Zhang, Weishuo Sun +7Dynamic Robotic ManipulationWorld Action Models