cs.RONov 16, 2026

ThinkWAM: Rethinking Latent Futures for Action Refinement

Authors: Haoyun Tang, Baqiao Deng, Enjin Zhou, Songwen Tian, Keyao Xu, Bowen Jing, Kun Wang, Ke Ma, 

Organizations: Xi’an Jiaotong University · Nankai University · The University of Manchester · Nanyang Technological University · 

Abstract

World-action models (WAMs) couple action generation with predictions of future states. An initial action chunk may be flawed. A future prediction that is not conditioned on the chunk may fail to reflect the changes that the proposed actions would induce in visual representations, limiting its usefulness for refinement before execution. We introduce ThinkWAM, a latent WAM that establishes bidirectional feedback before execution through a proposal-rethink-refine loop: the policy first proposes a complete action chunk, the proposal corrects multi-horizon future latents through a learned residual, and the corrected latents are injected into a second pass of the same action head, which generates the executable chunk. An affordance-guided observation-action memory and a near-fine, far-coarse multi-horizon prediction scheme, both motivated by predictive coding, support this loop, and all prediction stays in latent space. ThinkWAM achieves 98.70% success on LIBERO, 84.76% on LIBERO-Plus without OOD fine-tuning, 85.75%/65.85% on RoboTwin 2.0 Clean/Random, and 65.17% on RoboCasa24. ThinkWAM achieves 64.4% mean success across three real-world OOD conditions. The mean two-pass inference latency per action chunk is 153.48 ms on LIBERO, 164.61 ms on RoboTwin, and 101.2 ms on the Galaxea A1Z. Ablations show complementary gains from rethinking and affordance guidance. Diagnostics indicate that proposal-conditioned corrections change both predicted futures and action trajectories, showing that the corrected future representations improve the final action.

Figures & tables

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Latent Action as Intention Enables Efficient Future Imagination for World Action Models

    Aug 25, 2026Xiang Li, Yupeng Zheng, Songen Gu +11Latent Action LearningWorld Models for Robotics

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

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

  3. Foresight Without Seeing: Latent Futures for World Action Models

    Aug 12, 2026Jiakai Huang, Zhongbo Wu, Siyu Xu +5World Models for RoboticsVisuomotor Policy Learning