LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies

• URL: http://arxiv.org/abs/2606.15768v1
• Date: Sun, 14 Jun 2026 12:06:58 GMT
• Title: LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies
• Abstract summary: We present LaWAM, a latent-action-conditioned Latent World Model (LaWM)<n>LaWAM exposes predictive dynamics to robot policies through compact latent visual subgoals instead of reconstructed future video.<n>LaWAM runs in 187 ms per action-chunk prediction and achieves up to 24x lower wall-clock latency than pixel-space WAMs.
• Score: 31.88655699486955
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language-Action models (VLAs) leverage large-scale vision-language pretraining for semantic robot control, but often lack explicit foresight into how robot actions change the scene. World-Action Models (WAMs) address this limitation by conditioning policies on predicted futures, yet existing approaches typically rely on computationally expensive video generation with substantial pixel-level redundancy. We present LaWAM, a Latent World Action Model that exposes predictive dynamics to robot policies through compact latent visual subgoals instead of reconstructed future video. At the core of LaWAM is a latent-action-conditioned Latent World Model (LaWM). We obtain LaWM by training a latent action model in the latent space of a pretrained vision foundation model and repurposing its forward decoder to predict future observation features for scene evolution. LaWAM then conditions action generation on these predicted latent visual subgoals to enable dynamics-aware robot control. LaWAM achieves state-of-the-art or competitive success rates (SRs) across LIBERO (98.6% SR), RoboTwin (91.22% SR), and real-world manipulation tasks while retaining low-latency inference. LaWAM runs in 187 ms per action-chunk prediction and achieves up to 24x lower wall-clock latency than pixel-space WAMs.

Related papers

• LD4WAM: Learning Latent Dynamics from Human Videos for World Action Models [14.629670168049984]
  We propose motion-aligned latent dynamics as an embodiment-agnostic representation to bridge video priors and low-level actions.<n>We present LD4WAM, which pairs a Latent Dynamics Model trained with semantic reconstruction and real motion alignment with a World Dynamics Action Model built as a mixture-of-transformers (MoT)<n>Pretrained on our curated unified dataset of over 5,000 hours of human and robot data, LD4WAM performs strongly in RoboTwin simulation and on real robots equipped with both grippers and dexterous hands.
  arXiv  Detail & Related papers  (2026-08-23T13:09:30Z)
• DELE-w0.5: Inferring Action from Future Latent State for Robotic Manipulation [11.665741844323984]
  We argue that video generation is an unnecessary objective for world-action modeling.<n>For robotic manipulation, the goal of a world model is not to reproduce how the world looks at every intermediate moment, but to predict the state that the world will reach after an action is executed.<n>We propose DELE-w0.5, which infers robot actions from predicted future states without relying on video generation.
  arXiv  Detail & Related papers  (2026-08-22T18:17:36Z)
• GigaWorld-Policy-0.5: A Faster and Stronger WAM Empowered by AutoResearch [66.7936140772745]
  We present GigaWorld-Policy-0.5, an action-centered WAM designed for more efficient robot control.<n>For efficient inference, GigaWorld-Policy-0.5 introduces a Mixture-of-Transformers architecture that separates visual dynamics modeling and action generation into specialized experts.<n>Experiments and ablations show that GigaWorld-Policy-0.5 preserves the training benefits of future visual dynamics while improving inference efficiency for robot control.
  arXiv  Detail & Related papers  (2026-07-15T15:46:42Z)
• WALA Learning Executable Latent Actions from Action-Labeled Demonstrations and Action-Free Videos [38.959415159248046]
  WALA is a framework for learning executable latent actions from both action-labeled demonstrations and action-free videos.<n>We show that WALA achieves strong performance on RoboTwin, sets a new state-of-the-art result on RoboCasa with 75.2% average success.
  arXiv  Detail & Related papers  (2026-07-13T11:02:04Z)
• RotVLA: Rotational Latent Action for Vision-Language-Action Model [54.22746299071677]
  We introduce RotVLA, a VLA framework built on a continuous rotational latent action representation.<n>Latent actions are modeled as elements of SO(n), providing continuity, compositionality, and structured geometry aligned with real-world action dynamics.<n>RotVLA consists of a VLM backbone and a flow-matching action head, pretrained on large-scale cross-embodiment robotic datasets and human videos.
  arXiv  Detail & Related papers  (2026-05-13T11:58:02Z)
• From Imagined Futures to Executable Actions: Mixture of Latent Actions for Robot Manipulation [88.39072412680633]
  We propose MoLA, a control-oriented interface that transforms imagined future videos into executable representations.<n>We evaluate our approach on simulated benchmarks and real-world robot manipulation tasks.
  arXiv  Detail & Related papers  (2026-05-12T14:15:16Z)
• DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA [72.9197085473598]
  DIAL is a framework bridging high-level decision making and low-level motor execution through a differentiable latent intent bottleneck.<n>A VLM-based System-2 performs latent world modeling by latent visual foresight within the VLM's native feature space.<n>A lightweight System-1 policy then decodes this predicted intent together with the current observation into precise robot actions.
  arXiv  Detail & Related papers  (2026-03-31T15:02:27Z)
• Do World Action Models Generalize Better than VLAs? A Robustness Study [25.418384276142223]
  Vision-language-action (VLA) has achieved notable success across a variety of robotic tasks.<n>World action models (WAMs) are built upon world models that are trained on large corpora of video data to predict future states.<n>We evaluate their performance on the LIBERO-Plus and RoboTwin 2.0-Plus benchmarks under various visual and language perturbations.
  arXiv  Detail & Related papers  (2026-03-23T15:13:15Z)
• GigaWorld-Policy: An Efficient Action-Centered World--Action Model [50.107640832046464]
  GigaWorld-Policy is an action-centered WAM that learns 2D pixel-action dynamics while enabling efficient action decoding, with optional video generation.<n> Experimental results on real-world robotic platforms show that GigaWorld-Policy runs 9x faster than the leading WAM baseline, Motus.
  arXiv  Detail & Related papers  (2026-03-18T00:52:02Z)
• mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs [5.109732854501585]
  We introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations.<n>Our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.
  arXiv  Detail & Related papers  (2025-12-17T18:47:31Z)
• Latent Action Pretraining from Videos [156.88613023078778]
  We introduce Latent Action Pretraining for general Action models (LAPA)<n>LAPA is an unsupervised method for pretraining Vision-Language-Action (VLA) models without ground-truth robot action labels.<n>We propose a method to learn from internet-scale videos that do not have robot action labels.
  arXiv  Detail & Related papers  (2024-10-15T16:28:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.