UBTECH Introduces Thinker-WM, Reports First-Place Results on Three Embodied World Model Benchmarks

1. 16:00
   量子位

   UBTECH introduces Thinker-WM, unifying video, depth, and robot actions in one world model

   UBTECH introduces Thinker-WM, a multimodal world model that jointly models RGB video, depth, and robot actions through a shared Unified Dynamics Transformer. In material supplied by UBTECH and republished by QbitAI, the company reports leading results on WorldOlympiad, EWMBench, and WorldArena1.0 Track-1, including scores of 0.740 on WorldOlympiad and 4.66 on EWMBench. The model supports six capability modes through shared weights, and adding depth improved RoboTwin single-task success by 15 percentage points over a baseline without depth.

Trends compare the same continuously observed participants, which may cover fewer sources than current heat. Point or click to inspect hourly heat, or use left and right arrow keys.