Bridge-WA: Predicting Where and How the World Changes for Robotic Action
Paper • 2607.02195 • Published
How to use baiyu858/Bridge-WA-VLABench with Transformers:
# Load model directly
from transformers import WorldBridgeXVLA
model = WorldBridgeXVLA.from_pretrained("baiyu858/Bridge-WA-VLABench", trust_remote_code=True, dtype="auto")Bridge-WA is a lightweight world-action framework that distills a frozen future-change teacher into compact world-change priors (future tokens, change maps, and motion-flow maps) to guide robotic action generation.
If you find Bridge-WA useful in your research, please cite:
@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}
}