Bridge-WA-World-Teacher-BridgedataV2

Bridge-WA: Predicting Where and How the World Changes for Robotic Action

This repository contains the policy checkpoint for Bridge-WA, as presented in the paper Bridge-WA: Predicting Where and How the World Changes for Robotic Action.

• Project Page: https://hcplab-sysu.github.io/BRIDGE-WA
• Repository: https://github.com/HCPLab-SYSU/BRIDGE-WA

About Bridge-WA

Bridge-WA is a lightweight world-action framework that distills a frozen future-change teacher into three compact priors:

1. Future tokens for intended outcomes.
2. Change maps for intervention support.
3. Motion-flow maps for local transition direction.

A WorldBridge conditions the action transformer on these priors through multi-source attention memories and spatial-temporal biases, while the teacher model is removed at inference. By focusing action generation on where and how the scene will change, Bridge-WA suppresses nuisance appearance factors such as background, lighting, and distractors, leading to better generalization without deployment-time dense future-image generation.

Usage and Evaluation

For installation, training, and evaluation instructions on benchmarks such as VLABench, please refer to the official GitHub repository.

Citation

@article{bai2026bridgewa,
  title   = {Bridge-WA: Predicting Where and How the World Changes for Robotic Action},
  author  = {Bai, Yongjie and Wang, Hanting and Dai, Mingtong and Zhong, Qijun and Liu, Yang and Lin, Liang},
  year    = {2026}
}