cs.RONov 13, 2026

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

Authors: Yu Zhou, Changsong Lei, Yixiang Wei, Kairui Wen, Huankun Sheng, Ming Li, Yeying Fan, Yu-Hui Wen, 

Organizations: Tsinghua University · Beijing Jiaotong University

Abstract

World models that support planning, control, and interpretable reasoning require latent representations that genuinely encode action. Yet the large-scale video data from which such models are learned overwhelmingly lack action annotations. Latent-action models infer action-like variables from observation sequences, enabling action-aware learning from unlabeled videos or trajectories. A fundamental concern, however, remains: a latent that improves future prediction may simply encode the current state, episode progress, task identity, or other predictive shortcuts. Existing evaluation practices---reconstruction loss, rollout fidelity, action recovery, or downstream performance---cannot isolate this distinction. We introduce the Latent Action Support Ratio (LASR), a dimensionless metric that measures, under offline evaluation on held-out data, whether a learned latent provides predictive support for action-related targets beyond the current state and non-action alternatives. LASR provides two complementary summaries: GeoLASR captures typical support, while WorstLASR identifies the weakest case. We instantiate the LASR benchmark with recorded actions and corresponding physical state changes as targets, episode-held-out splits, one predictor type for all comparisons, and positive and negative controls. Experiments on recorded robot manipulation data and a matched multi-task panel demonstrate that LASR detects known action information, rejects representative non-action shortcuts, and enables graded comparisons among imperfect latent representations. We further use LASR to train and audit a lightweight residual adapter, obtaining descriptively better point estimates whose uncertainty is assessed with paired comparisons. LASR thus provides a principled offline protocol for measuring action-related predictive support in world-model latents.

Figures & tables

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

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

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

  2. What Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction

    Jun 5, 2026Jewon Yeom, Hanseul Kim, Jeongjae Park +3Video Representation LearningPredictive Representation Learning

  3. OneLA: Learning Many Worlds from Latent Actions

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