门控后见蒸馏：利用未来截图提升移动 GUI 代理训练

AI 摘要

Gated Hindsight Distillation (GHD) 是一种改进移动 GUI 代理训练的方法，利用未来截图作为特权信息，在标准模仿失败时恢复正确的推理。该方法在 AndroidWorld 和 AndroidLab 上，通过两种视觉语言模型，相比 GRPO 提高了任务成功率。代码和检查点将公开。

正文节选

The Next Screenshot Knows: Gated Hindsight Distillation for Mobile GUI Agents Abstract Gated Hindsight Distillation improves GUI agent training by using future screenshots as privileged evidence to recover correct reasoning when standard imitation fails. GUI agents are commonly trained offline from successful interaction trajectories. Standard training decomposes each trajectory into prefix-action pairs: the agent predicts an action from the current screen and interaction history, while the subs

发布时间：—

抓取时间：2026-08-12 08:51

来源机构：Hugging Face