环境地图：面向长周期智能体的结构化环境表示
Environment Maps: Structured Environmental Representations for Long-Horizon Agents
AI总结：
本文提出环境地图，通过整合异构证据构建结构化图表示，提升长周期任务的鲁棒性。在WebArena基准上，环境地图使智能体成功率提升至28.2%，优于基线方法。
摘要：
尽管大语言模型（LLMs）发展迅速，复杂软件工作流的可靠自动化仍是一个开放问题。在长周期设置中，智能体常面临级联错误和环境随机性；动态界面中的单个失误可能导致任务失败，产生幻觉或试错。本文提出环境地图：一种持久、智能体无关的表示，通过整合屏幕录制和执行轨迹等异构证据，构建结构化图。该表示包含四个核心组件：（1）上下文（抽象位置）、（2）动作（参数化 affordances）、（3）工作流（观察到的轨迹）、（4）隐性知识（领域定义和可重用过程）。我们在WebArena基准的五个领域评估了该框架。配备环境地图的智能体成功率达到28.2%，几乎翻倍于仅限于会话绑定上下文的基线（14.2%），并优于能访问生成环境地图所用原始轨迹数据的智能体（23.3%）。通过为模型和环境提供结构化接口，环境地图建立了持久的基础，支持长周期规划，具有可解释、可编辑和渐进可改进的特性。
英文摘要：
Although large language models (LLMs) have advanced rapidly, robust automation of complex software workflows remains an open problem. In long-horizon settings, agents frequently suffer from cascading errors and environmental stochasticity; a single misstep in a dynamic interface can lead to task failure, resulting in hallucinations or trial-and-error. This paper introduces $\textit{Environment Maps}$: a persistent, agent-agnostic representation that mitigates these failures by consolidating heterogeneous evidence, such as screen recordings and execution traces, into a structured graph. The representation consists of four core components: (1) Contexts (abstracted locations), (2) Actions (parameterized affordances), (3) Workflows (observed trajectories), and (4) Tacit Knowledge (domain definitions and reusable procedures). We evaluate this framework on the WebArena benchmark across five domains. Agents equipped with environment maps achieve a 28.2% success rate, nearly doubling the performance of baselines limited to session-bound context (14.2%) and outperforming agents that have access to the raw trajectory data used to generate the environment maps (23.3%). By providing a structured interface between the model and the environment, Environment Maps establish a persistent foundation for long-horizon planning that is human-interpretable, editable, and incrementally refinable.
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。
