HiconAgent：面向 GUI 智能体的历史上下文感知策略优化
HiconAgent: History Context-aware Policy Optimization for GUI Agents
把「历史观测该保留多少」变成策略优化里的显式变量：DCS 在采样阶段给智能体不同长度的历史，AHC 用双分支（压缩分支丢观测、保留动作作为信息流锚点）加对齐损失约束两分支一致，这套「变长历史采样 + 压缩分支对齐」可直接迁移到视频/多帧输入的视觉决策模型，用来在长时序视觉上下文下省算力。
原文依据 Dynamic Context Sampling (DCS) presents the agent with variable-length histories during sampling
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论图形用户界面（GUI）智能体需要有效利用历史上下文来完成序列导航任务。引入过去的动作与观测可以显著改善决策，但朴素地使用完整历史会带来过高的计算开销，并可能被无关信息干扰。在本工作中，我们提出 HiconAgent，一种以历史上下文感知策略优化（HCPO）训练的 GUI 智能体，用于高效且有效地利用历史信息。HCPO 通过在采样与策略更新两个环节中显式优化历史的使用，集成了两个互补组件：（1）动态上下文采样（DCS）在采样阶段向智能体呈现变长历史，使其能够自适应地使用最相关的历史上下文，从而提升序列决策质量；（2）锚点引导的历史压缩（AHC）通过双分支优化策略改进策略更新阶段，其中压缩分支丢弃历史观测，但保留历史动作作为信息流锚点。压缩分支与非压缩分支通过历史增强的对齐损失进行耦合，以强制二者对历史的使用保持一致，从而在性能损失极小的前提下实现效率提升。在主流 GUI 导航基准上的大量实验表明，我们的模型具有强劲性能。尽管规模更小，HiconAgent-3B 在 GUI-Odyssey 上以 +8.46% 的视觉定位和 +11.32% 的单步成功率超过 GUI-R1-7B，同时在 AndroidControl 与 AITW 上取得可比结果，并带来最高 2.47 倍的计算加速与 60% 的浮点运算量降低。
英文原文摘要 1706 字符
Graphical User Interface (GUI) agents require effective utilization of historical context to perform sequential navigation tasks. While incorporating past actions and observations can significantly improve decision-making, naively using full history leads to excessive computational overhead and potential distraction from irrelevant information. In this work, we introduce ****HiconAgent****, a GUI agent trained with ****History Context-aware Policy Optimization (HCPO)**** for effective and efficient utilization of historical information. HCPO explicitly optimizes history usage in both sampling and policy updates by integrating two complementary components: ****(1) Dynamic Context Sampling (DCS)**** presents the agent with variable-length histories during sampling, enabling adaptive use of the most relevant historical context to improve sequential decision quality; ****(2) Anchor-guided History Compression (AHC)**** refines the policy update phase via a dual-branch optimization strategy, where the compressed branch drops history observations while keeping history actions as information flow anchors. The compressed and uncompressed branches are coupled through a history-enhanced alignment loss to enforce consistent history usage, achieving efficiency with minimal performance degradation. Extensive experiments on mainstream GUI navigation benchmarks demonstrate the strong performance of our model. Despite its smaller size, HiconAgent-3B outperforms GUI-R1-7B by ****+8.46****% grounding and ****+11.32****% step successful rate on GUI-Odyssey, while achieving comparable results on AndroidControl and AITW, with up to ****2.47x computational speedup**** and ****60% FLOPs reduction****.