Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization

1 sources · 2 reports·First 2026-08-27 12:00→Latest 2026-08-28 12:00