ForkCast：基于反事实未来的视觉-语言-动作策略的预测性恢复
ForkCast: Predictive Recovery for Vision-Language-Action Policies via Counterfactual Futures
AI总结：
ForkCast是一种预测性恢复框架，通过比较反事实未来确定是否及如何干预冻结的VLA策略，在模拟RoboTwin 2.0任务中大幅提升了操作成功率，还可迁移适配至真实机器人场景。
摘要：
视觉-语言-动作（VLA）策略利用相机图像和语言指令执行操作任务，但物体打滑或视觉遮挡会干扰执行过程。恢复通常在策略训练期间学习，或委托给监控器，该监控器在检测到故障后触发脚本式修复。我们提出ForkCast，一种预测性恢复框架，它通过比较反事实未来来确定是否以及如何干预冻结的VLA策略。在每个决策点，ForkCast评估三个候选方案：执行策略的标称动作块（Commit）、调整相机视角（Reobserve）和重新规划机械臂轨迹（Replan）。一个基于候选方案的世界模型在共同的未来时间范围内预测每个候选方案的潜在结果，而一个共享的任务进度头对预测结果进行评分。通过将策略自身的延续与修复候选方案一起评估，基于任务进度的单一比较即可确定是否以及如何进行干预，无需故障检测器、干预阈值或更新VLA主干网络。在五个注入了遮挡和物体打滑的模拟RoboTwin 2.0任务上，ForkCast将合并成功率从冻结π₀.₅策略的13.3%提升至30.0%，且无需更新ForkCast即可迁移到另外两个VLA主干网络。在真实机器人候选滚动输出上进行适配后，ForkCast还在干净和可见性受限的条件下均比基础策略提高了成功率。代码将发布以造福研究界。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
