WMPO: World Model-based Policy Optimization for Vision-Language-Action Models

作者：Fangqi Zhu, Zhengyang Yan, Zicong Hong, Quanxin Shou, Xiao Ma, Song Guo · 单位：Hong Kong University of Science and Technology ByteDance Seed · 会议/期刊：ICLR 2026 · 日期：2025-11-12 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Auxiliary Tasks - World Model & Visual Prediction / World Model