ActionPiece rethinks action tokenization for VLA models, hits 94.8% on LIBERO
burny_tech · x · 2026-09-20
The paper ActionPiece: Rethinking Action Tokenization for VLA Models redesigns action tokenization for vision-language-action models, preserving physical relationships between actions via Physical Rank Consistency (PRC) instead of naive discretization.
- 94.8% on LIBERO
- 68.8% on the out-of-distribution LIBERO-Plus
The key idea is encoding physical constraints directly into the tokenizer to improve robot policy generalization.
More from Embodied
- Open-source dual-arm robot solves tangram puzzles with 90 demos, ~60% success — DanGrover · 2026-11-12
- Robotics has three regional playbooks: China volume, Europe inspection, US pilots — ingliguori · 2026-11-12
- 3,500 Laser-Armed AI Submarine Robots Patrol Norwegian Salmon Farms — philfung · 2026-11-12
- Model Autonomously Designs a Surgical-Robot-Style Robotic Arm CAD After 75 Minutes — _Stocko_ · 2026-11-12
- Dev releases local world model turning any image into interactive video, runs on RTX 3060 — lucidml_lover · 2026-11-12
- RoboRisk benchmark shows foundation models pose high harm rates in robotics; LEASH harness cuts risk — UoLiverpool · 2026-11-12