NTU's Facet-0 robot model lifts precision assembly success from 16% to 82%
jiqizhixin · x · 2026-09-11
NTU's PINE Lab unveiled Facet-0, a robot foundation model for contact-rich precision assembly. Existing models struggle badly: π0.5 averages 10% success and GR00T N1.7 just 4% on insertions with 0.10–0.30mm clearances. Facet-0 judges whether it's performing correctly from real contact feedback and improves from failure experience, with the training pipeline lifting success from 16% (pretraining) to 38% (RL post-training) to 82% (lightweight adaptation), handling 23 sub-goals per task across alignment, insertion, pressing, and locking.
More from Embodied
- Figure crosses 86,000 weekly active users uploading data for robotics dataset — adcock_brett · 2026-09-11
- OpenArm: Tokyo startup enactic open-sources a humanoid robot with full CAD, firmware and sim tooling — lukas_m_ziegler · 2026-09-11
- Leaked GPT-6 Astra Scores 46% vs 12% for MolmoAct2 on Bimanual Robot Tasks — damianplayer · 2026-09-11
- UMR: open-source unified motion retargeting for humanoids via learned point-cloud correspondence — ChongZzZhang · 2026-09-11
- Autonomous Lamp open-source companion robot adds dust, air, temperature and humidity sensing — dee_hw · 2026-09-11
- Cybertruck is Tesla's rolling engineering lab: 48V, steer-by-wire, Etherloop — XFreeze · 2026-09-11