Event
X-Era AI Tuoyuan Zhihui Open-Sources 4D World Action Model E0.5
3 reports · 3 sources
X-Era AI Tuoyuan Zhihui has open-sourced E0.5, a native 4D world action model that uses a 3D foundation model as its backbone, incorporates future 3D geometry prediction into action learning, and performs embodied pretraining on over 160,000 real robot trajectories. The team reports that on the AgileX Piper dual-arm robot and the Zerith H1 humanoid robot, E0.5 achieved overall average success rates of 85.0% and 88.3%, respectively, 13.3 and 26.7 percentage points higher than π0.5; out-of-distribution average success rates were 73.3% and 76.7%. Before pretraining, the two platforms' average success rates across in-distribution and out-of-distribution settings were 70.0% and 81.7%, improving by about 15.0 and 6.7 percentage points after pretraining. The real-robot evaluation tasks were mouse repositioning, charger storage, and trash cleanup, with each policy accumulating 120 trials across the two platforms, for a total of 480 trials. On RoboTwin 2.0, without embodied data pretraining, E0.5 achieved success rates of 89.90% and 50.09% under clean and randomized settings, respectively.
Developments
- X-Era AI Tuoyuan Zhihui releases open-source 4D world action model E0.5, with out-of-distribution real-robot success rate surpassing π0.5 · 2 reports
- E0.5: A Geometry-Native World-Action Model Built on a 3D Foundation Model Backbone · 1 reports
Report timeline
- X-Era AI Tuoyuan Zhihui releases open-source 4D world action model E0.5, with out-of-distribution real-robot success rate surpassing π0.5
X-Era AI Tuoyuan Zhihui released the open-source native 4D world action model E0.5, which uses a 3D foundation model as its backbone and integrates future 3D geometry prediction directly into action learning. On the AgileX Piper dual-arm robot and the Zerith H1 humanoid robot, after pretraining on about 160,000 robot trajectories, E0.5 achieved overall average success rates of 85.0% and 88.3%, respectively, improving on π0.5 by 13.3 and 26.7 percentage points; its out-of-distribution average success rates were 73.3% and 76.7%.
OFweek Robotics ·
- X-Era AI opensources 4D world action model E0.5, real-robot success rate surpasses π0.5
X-Era AI has open-sourced its native 4D world action model E0.5, achieving overall average success rates of 85.0% and 88.3% on the AgileX Piper dual-arm robot and the Zerith H1 humanoid robot, respectively, 13.3 and 26.7 percentage points higher than π0.5.
机器之心 · 微信公众号 ·
- E0.5: A Geometry-Native World-Action Model Built on a 3D Foundation Model Backbone
The Era AI Lab team proposes E0.5, a geometry-native world-action model built directly on 3D foundation models that treats 3D geometry as the native representation for world modeling and action prediction, and explicitly predicts future 3D states and actions.
arXiv cs.RO ·