外观
论文arXiv 2610.1792215 天前
WoVR+：作为可靠模拟器的世界模型，用于通过 RL 优化多用途 VLA 策略
原文链接：WoVR+: World Model as Reliable Simulator for Versatile VLA Policy Optimization with RL（arXiv 2610.17922 · 发布日期：2026-10-13）
WoVR+: World Model as Reliable Simulator for Versatile VLA Policy Optimization with RL
WoVR+通过幻觉感知世界模型与分阶段想象训练，减少 VLA-RL 真机交互，并在 RoboTwin 和6项真机任务上验证提升。
核心要点
- 动作条件世界模型可降低 VLA-RL 真机交互成本，但幻觉会误导策略优化。
- WoVR+重设计学习环境和想象交互，提供多视角预测、本体更新并缓解幻觉。
- 分块噪声维持流式策略探索并限制动作抖动，分阶段本地训练应对长时程任务。
- 相比 SFT，RoboTwin 和6项真机任务成功率提升；长时程手机包装由30%升至75%。
验证范围与局限：验证范围为 RoboTwin、6项真机任务，以及长时程手机包装任务。
英文摘要 (Abstract)
Action-conditioned world models can reduce costly real-robot interaction in reinforcement learning for Vision-Language-Action models (VLA-RL), but hallucinations can mislead policy optimization. Extending them to richer VLA interfaces and long-horizon tasks raises challenges in spatiotemporal and cross-modal consistency, exploration-reliability balance in RL, and compounding rollout errors. We present WoVR+, extending WoVR's hallucination-aware WM-RL framework by redesigning the learned environment and imagined interaction protocol: A VLA-interface-aligned world model provides multi-view prediction and proprioceptive updates while mitigating hallucinations. Chunk-wise noise preserves flow-based policy exploration while limiting prediction-degrading action jitter. For long-horizon tasks, stage-local imagined training combines stage-entry resets, stage-specific rewards, local horizons, and controllable sampling to shorten generated prefixes and balance training across stages. WoVR+ improves success over SFT by 21.09% and 16.7% on RoboTwin and 6 real-robot tasks. Long-horizon phone-packaging success rises from 30% to 75%. Videos are available at https://youmo445.github.io/WoVR_Plus.github.io/.
作者：Shangqing Zhou, Zhennan Jiang, Ning Ma, Xin Fu, Yupeng Zheng, Shuai Tian, Huangrui Li, Jiahao Liu, Chao Yu, Zhongpu Xia
来源
本页事实对应的公开页面，以原文为准。