# BAAI releases multimodal representation world model Wujie·RoboBrain Orca

- Source: 智源研究院 · 微信公众号
- Published: 2026-07-07T05:06:22.000Z
- RoboHot: https://robohot.news/en/items/w7j55f4hk0l78y47ylrz6oa4z
- Original: https://mp.weixin.qq.com/s?__biz=MzI2MDcxMzQzOA==&mid=2247550800&idx=1&sn=f91ad6a9a7e29e4acad25f54cdea5ac8&chksm=eb65fef2d32f0fb92e972b6f1c7f4c61ff37c945471681d80ffde7bc6a2f5d1a3f0a9e5f7c4c&scene=126&sessionid=1792575143#rd

## Summary

BAAI has released the multimodal representation world model Wujie·RoboBrain Orca. Its core idea is to shift from predicting the next word, next frame, or next action to predicting the next world state: it first learns a unified latent world representation, then adapts to downstream tasks through three lightweight readout modules for text, images, and actions.

## Why it matters

The original text explains the training paradigm shift from predicting the next output to predicting the next state, as well as the approach of a frozen backbone plus lightweight readout modules, which is transferable to readers following the world model route.