Long-WAM: Scaling the Context of World-Action Models

Key points

• Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action.1
• We first learn causal prediction from robot and egocentric videos without action labels, then preserve this history-to-future structure during world-action adaptation.1
• On RoboCasa GR-1, increasing context from 0.0 to 19.2 seconds raises success from 63.3% to 78.7%, whereas a bidirectionally pretrained initialization shows no net gain; robot-domain AR pretraining further raises peak success on GR-1 and LIBERO-Long.1
• As a memory-informed executor, Long-WAM also complements higher-level planning in composite tasks.1

Sources (1)

• [1]Long-WAM: Scaling the Context of World-Action Models
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 05:58 PM

  We present Long-WAM, a model-system framework for scaling the context of causal world-action models under real-time control constraints.

  Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action.

Extractive summary: sentences quoted from the sources.