具身开发者社区

外观

论文arXiv 2610.1865015 天前

ForkCast：通过反事实未来为 Vision-Language-Action 策略进行预测性恢复 ​

ForkCast通过比较反事实未来，为冻结VLA策略预测何时及如何恢复，并在仿真RoboTwin 2.0和真实机器人上验证。

AI 整理
  • 本体 SDK 与控制
  • 具身模型

原文链接：ForkCast: Predictive Recovery for Vision-Language-Action Policies via Counterfactual Futures（arXiv 2610.18650 · 发布日期：2026-10-13）

ForkCast: Predictive Recovery for Vision-Language-Action Policies via Counterfactual Futures

ForkCast通过比较反事实未来，为冻结VLA策略预测何时及如何恢复，并在仿真RoboTwin 2.0和真实机器人上验证。

核心要点 ​

  • VLA策略易受物体滑移或视觉遮挡干扰，现有恢复常依赖训练内学习或故障检测后脚本修复。
  • ForkCast用候选条件世界模型预测共同未来结果，并以共享任务进度头评分候选。
  • 在五个带遮挡和滑移的仿真RoboTwin 2.0任务中，成功率由13.3%提升至30.0%。
  • 无需更新ForkCast即可迁移到两个额外VLA骨干；真实机器人适配后也优于基础策略，代码将发布。

验证范围与局限：验证范围包括五个注入遮挡和物体滑移的仿真RoboTwin 2.0任务，以及经真实机器人候选轨迹适配后的干净和可视受限条件。

英文摘要 (Abstract) ​

Vision-language-action (VLA) policies perform manipulation tasks using camera images and language instructions, but object slip or visual occlusion can disrupt execution. Recovery is typically learned during policy training or delegated to a monitor that triggers a scripted repair after detecting a failure. We introduce ForkCast, a predictive recovery framework that compares counterfactual futures to determine whether and how to intervene in a frozen VLA policy. At each decision point, ForkCast evaluates three candidates: executing the policy's nominal action chunk (Commit), adjusting the camera viewpoint (Reobserve), and replanning the arm trajectory (Replan). A candidate-conditioned world model forecasts the latent outcome of each candidate at a common future horizon, and a shared task-progress head scores the predicted outcomes. By evaluating the policy's own continuation alongside the repair candidates, a single comparison based on task progress determines both whether and how to intervene, without a failure detector, an intervention threshold, or updates to the VLA backbone. On five simulated RoboTwin 2.0 tasks with injected occlusion and object slip, ForkCast increases pooled success from 13.3% for the frozen $π_{0.5}$ policy to 30.0% and transfers to two additional VLA backbones without updating ForkCast. After adaptation on real-robot candidate rollouts, ForkCast also improves success over the base policy under both clean and visibility-limited conditions. Code will be released to benefit the research community.

作者：Chenru Wen, Huajian Zeng, Haokai Ding, Jian Kang, Anqing Duan, Dawei Zhou, Dezhen Song, Xingxing Zuo

来源

本页事实对应的公开页面，以原文为准。

  1. arXiv 2610.18650：ForkCast: Predictive Recovery for Vision-Language-Action Policies via Counterfactual Futures访问 2026-10-21