UBTECH proposes Thinker-WM embodied world model
What happened
UBTECH proposes the embodied world model Thinker-WM, which encodes RGB video, metric depth and robot actions separately as tokens in a unified latent space, then processes them with a shared Unified Dynamics Transfor. Currently, the publicly available material provides only an abstract and does not give experimental settings or results, so the model's actual performance cannot yet be judged.
AI 根据报道生成 · 3 days ago更新
Reporting timeline
Follow the reporting to explore different perspectives.
- GuyuehomeUBTECH proposes Thinker-WM unified world model with shared multimodal backbone
UBTECH proposes the embodied world model Thinker-WM, which encodes RGB video, metric depth, and robot actions separately as tokens in a unified latent space, processed by a shared Unified Dynamics Transfor. The material provides only an abstract and does not give experimental setup or results.
Activity for this event
- Current
- 3
- Peak
- 10Nov 8 03:00
- Past 24 hours
- -50%
The trend compares participants observed throughout the same period, so its range may be smaller than the current interest count. Move or tap to inspect each hour; use the arrow keys on a keyboard.