Latest / Chips, compute and energy
GreenWAM world foundation model covers 20 robot families for prediction and control
See how this story connectsGennady Sidorov, Valerii Kornilov, Denis Grankin, Polina Fedotova and Aleksey Postnikov submitted GreenWAM, a world foundation model for robotics, to arXiv on 13 October 2026. It uses a shared anatomical action space that describes 20 robot families, from bimanual arms to dexterous humanoids, and is trained jointly in visual, action-conditioned and world-action modes. The authors report that it scores among the top entries on the RoboTwin 2.0 leaderboard, reaches the highest zero-shot LIBERO-plus success among the compared methods, and was adapted to a real humanoid.
Why it matters
The paper reports one model working as both a world model and a policy across 20 robot families, with results on a real humanoid.
- When
- 13 Oct 2026
- ID
WTA-N-2026-0297- Where
- International
Summary by wheresthe.ai, from Hugging Face daily papers. Checked against the source on 14 Oct 2026.
Source copy: download the page as saved on 18 Oct 2026 (PDF). Also held by the Internet Archive.
Cite this story
WTA-N-2026-0297: “GreenWAM world foundation model covers 20 robot families for prediction and control.” wheresthe.ai, 13 Oct 2026. https://wheresthe.ai/d/intl-greenwam-world-foundation-model-covers-20-robot-familie-2026-10/ (source: Hugging Face daily papers).
Spotted an error? Email [email protected].