论文

PACT 训练框架修复 GUI 智能体的计划-动作执行偏差

PACT: Plan-Action Consistency Training by Preserving Correct Plans and Repairing Actions in GUI Agents

精选理由

做 GUI 智能体训练的可以看看这篇:它发现模型想得对但动作做错的老问题,用屏蔽计划区间的负梯度和偏好修复来解决,AndroidWorld 成功率直接提了 8.63 个百分点,移动端和桌面端都适用。

GUI 智能体常出现推理中给出正确计划、执行动作却偏离计划的情况,一个错误动作会让多步任务整体失败。PACT(Plan-Action Consistency Training)针对这一执行缺口设计:先用带标注的参考动作定位执行错误,在计划 token 区间上屏蔽负的策略梯度贡献,保护已验证的计划;再以生成的推理作为固定上下文,通过带 stop-gradient 权重的偏好损失,让参考动作优于错误动作,且不改动原有的标量动作奖励。

正文

实验数据上,相比仅用结果奖励的基线,PACT 将 AndroidWorld 任务成功率从 60.34% 提升到 68.97%(+8.63 个百分点),AndroidControl High 上的规则化计划-动作一致性从 83.71% 提升到 89.74%(+6.03 个百分点);迁移到桌面端后,在 369 个任务的 OSWorld 基准上任务成功率从 40.92% 提升到 44.17%(+3.25 个百分点)。该方法解决的是 outcome-supervised RL 中把负优势广播到全部 token、误伤正确规划的信用分配冲突,对训练 GUI 智能体的工程团队有直接参考价值。

原文 · arXiv cs.AI

PACT: Plan-Action Consistency Training by Preserving Correct Plans and Repairing Actions in GUI Agents

GUI agents can formulate a correct immediate plan in their reasoning yet emit an action that fails to execute it, creating an execution gap. This recurring mismatch undermines reliable task completion, as a single execution error can derail a multi-step workflow. In outcome-supervised reinforcement learning, broadcasting a negative response-level advantage across all generated tokens risks penalizing valid planning together with erroneous execution. To address this credit-assignment conflict, we introduce PACT (Plan-Action Consistency Training), a training framework that combines reference-aligned plan protection and plan-aligned action repair without altering the original scalar action reward. Using annotated reference actions to identify execution gaps, PACT protects the verified plan by masking negative policy-gradient contributions across its token span. Keeping the generated reasoning unchanged as context, it then favors the reference action over the emitted erroneous action through a preference surrogate with stop-gradient pair weights. Compared with the outcome-only baseline, PACT raises AndroidWorld task success from 60.34% to 68.97% (+8.63 pp), while improving rule-based plan-action consistency on AndroidControl High from 83.71% to 89.74% (+6.03 pp). Adapting PACT to desktop agents also improves task success from 40.92% to 44.17% (+3.25 pp) on the full 369-task OSWorld benchmark. Together, these results establish PACT as an effective credit-assignment framework for bridging execution gaps and improving task completion across mobile and desktop GUI environments.

产业坐标与信源核验

本条情报归属 AITOP「论文」核心栏目,重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:GUI智能体、PACT、AndroidWorld、OSWorld,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Zhuohan Cai, Zhangxuan Gu, Runze Li, Beitong Zhou, Ming Wen, Hongbo Yu, Shuheng Shen, Changhua Meng),经 AITOP 多维评分管线结构化提炼。