PSG-JEPA Grounds Robot World Models with Physical State Awareness
zhenjun_zhao · x · 2026-08-13
To address the lack of reliable physical state identifiability in forward prediction objectives of JEPA world models, researchers proposed PSG-JEPA (Physical State Grounding JEPA).
- The Problem: Current JEPA-based world models learn action-conditioned predictive latent dynamics but fail to explicitly enforce reliable identifiability of robot-centric physical states from individual latents, limiting downstream planning and policy performance.
- The Method: PSG-JEPA introduces two complementary grounding objectives beyond forward prediction: grounding individual latents in robot proprioceptive states, and grounding latent pairs in multi-horizon joint-angle changes. These objectives are applied only during training, leaving inference architecture and computational cost unchanged.
- Results: Across three evaluation levels (latent identifiability via probing, goal-conditioned planning on frozen latents, and policy learning in simulation and on a real robot), PSG-JEPA consistently outperformed state-of-the-art latent world-model baselines.
More from Embodied
- Robot shows up at Costco to fetch the famous $1.50 hotdog — majidmanzarpour · 2026-10-16
- 15 Leju Kuavo humanoid robots run 2 months at Hangzhou auto-parts line, 98% success rate — CyberRobooo · 2026-10-16
- Baidu to give 1,000+ Hong Kong residents free robotaxi rides in city-first public trial — Baidu_Inc · 2026-10-16
- Auki demos spatial harness: robots and devices share spatial context via AR — broodsugar · 2026-10-16
- Tesla's 2-DOF car becomes 6-DOF flying demo — easier control problem than humanoids — micoolcho · 2026-10-16
- XREAL Aura, Android XR glasses powered by Gemini, named a TIME Best Invention of 2026 — DuRuofei · 2026-10-16