HKUST-GZ and EulerWanxiang open-source UniWAM, first to reveal human-robot co-training scaling
jiqizhixin · x · 2026-10-21
EulerWanxiang and HKUST (Guangzhou) released UniWAM, the open-source version of Ola-Sem—a Unified World-Action Model combining semantic understanding, future visual prediction, and action generation in one architecture. It is the first to reveal scaling trends of human-robot co-training, linking human egocentric data with robot data via shared physical-language supervision. It sets SOTA on LIBERO (99.2% avg success), LIBERO-Plus (92.6%), and RoboTwin 2.0 C2R (68.32%), with real-robot tests confirming instruction following, OOD generalization, and long-horizon capability.
More from Embodied
- MIT CSAIL's X-Hinges 3D-Prints Everyday Objects With Built-In Sensing Across Three Motion Axes — MIT News AI · 2026-10-23
- Formic raises $50M Series B to scale physical AI across 100+ US factories — garrytan · 2026-10-23
- fal starts fine-tuning a physical robot to draw, results next week — OdinLovis · 2026-10-23
- PEARL demo: an LLM agent that checks experiment results, revises plans, and re-tests — KevinKaichuang · 2026-10-23
- CU Boulder's ATLAS Institute Opens Tenure-Track Robotics Professorship — ryosuzk · 2026-10-23
- MiTaS multi-resolution tactile imitation learning hits 80% success vs 35.5% vision-only baseline — Jan_R_Peters · 2026-10-23