cs.LGNov 18, 2026

Latent Action Policies from Suboptimal Data

Authors: Donghyeon Ki, Xiaoyu Chen, Jun Ho Seo, Hayeong Lee, Kyungyoon Kim, Byung-Jun Lee, Li Zhao

Organizations: Korea University · Microsoft Research Asia

Abstract

Latent Action Models (LAMs) learn control representations from action-free video, but existing pipelines typically train latent policies with behavioral cloning, without an explicit policy improvement objective. Although offline RL offers a natural approach to policy improvement, standard methods do not transfer directly to learned latent action spaces. We find that the structure of learned latent actions makes naive offline RL unreliable: directly maximizing value over latent actions or fitting a unimodal policy can produce latent actions that cannot be realized by physical execution, while the latent representation can merge physical controls that receive different rewards. To understand when latent-space optimization remains valid, we derive conditions under which policy improvement transfers to decoded execution and under which latent and physical optima align. Building on these insights, we propose LAPIS for goal-conditioned latent policy improvement, grounding value-based policy improvement in observed state-latent transitions and using a conditional flow policy to capture their multimodal structure. LAPIS outperforms baselines on most tasks, with the clearest gains in visual manipulation, and remains effective even with only 1% action labels, where competing methods largely collapse.

Figures & tables

Appendix figures & tables19 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

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

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

  2. What Matters for Latent Actions in Robot Learning

    Aug 20, 2026Xizhou Bu, Qingda Hu, Lei Zhou +13Latent Action LearningRobot Learning

  3. Why Latent Actions Fail, and How to Prevent It

    May 13, 2026Jung Min Lee, Taehyun Cho, Li Zhao +1Latent Action LearningDisentangled Representation Learning