ACT-2 Preview claims 99% zero-shot success in unseen homes after one example
tonyzzhao · x · 2026-07-21
- ACT-2 Preview is presented as a robotics model that aims to unify broad generalization with high reliability.
- The claim: a single fine-tuning example can teach Memo a new behavior that then generalizes.
- The post says the system achieved a 99% success rate in zero-shot tests on real unseen homes.
More from Embodied
- Mundo turns a few room photos into an explorable, simulatable 3D world — at 1.5 FPS — mark_k · 2026-10-27
- Zurich Physical AI Hackathon puts attendees in a room with $1.5M of humanoids — ___Mufasaa · 2026-10-27
- Apple responds after iPhone Duo inner displays crack along the hinge — mobileraj · 2026-10-27
- VLA models encode task concepts as curved multi-dimensional geometry, 19-author study finds — ninamiolane · 2026-10-27
- Munro Live Posts Full Cybercab Engineering Deep-Dive Video — yunta_tsai · 2026-10-27
- Tesla's Robotaxi Fleet Keeps Growing in Austin — yunta_tsai · 2026-10-27