RoboScript's World In-Context Learning Hits 83.4% Success with Single Demo

HUST and ArcheOne AI propose World In-Context Learning (WICL) with the RoboScript infrastructure, letting robots learn manipulation from a single demonstration plus simulation rehearsal. It achieves 83.4% success across all 50 RoboTwin tasks, beating trained policies by about 20 points.

2026-11-11 ~ 2026-11-12 · 2 related posts

• RoboScript hits 83.4% success on RoboTwin 2.0 with one demo per task, beating trained policies by 20+ points — H-EmbodVis · 2026-11-11
• WICL: robots learn manipulation in-context, 83.4% success on all 50 RoboTwin tasks — _akhaliq · 2026-11-12