从智能体痕迹到信任：LLM智能体中的证据追踪与执行溯源
From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
AI总结：
本文系统综述了LLM智能体中的证据追踪与执行溯源方法，通过统一溯源视角连接检索、工具使用、记忆等环节，提出分类体系并讨论开放挑战。
摘要：
基于大语言模型（LLM）的智能体通过与外部工具、检索系统、记忆模块、环境及其他智能体交互，日益解决复杂任务。这些能力增强了智能体的自主性，但也使其行为更难以验证、调试和审计。仅凭最终答案的准确性无法解释输出是如何产生的、每个主张由哪些证据支持、工具调用是否合理、记忆如何影响后续决策或执行失败的根源。证据追踪和执行溯源通过建模检索到的证据、工具输出、记忆项、环境观察、中间主张、动作和最终答案在智能体执行过程中的连接方式，弥补了这一空白。本综述对LLM智能体中的证据追踪和执行溯源进行了系统回顾和概念框架构建。我们围绕统一的溯源视角组织相关工作，该视角连接了检索依据、主张支持、工具使用安全、记忆谱系、可观测性、调试、审计和恢复。我们引入了一个分类体系，涵盖追踪来源、证据和执行单元、溯源关系、追踪粒度和时机、表示形式以及信任功能。我们回顾了关键方法论方向，包括溯源表示、证据归因、工具使用溯源、运行时护栏、携带溯源的记忆、基于痕迹的可观测性和故障诊断。我们还绘制了现有基准、数据集和评估指标与溯源相关能力的映射，并讨论了评估如何从最终答案正确性转向过程级问责。最后，我们概述了开放挑战，包括统一痕迹模式、主张级和语义溯源、溯源感知的安全机制、现实执行痕迹基准、面向恢复的评估以及隐私感知的审计基础设施。
英文摘要：
Large language model (LLM)-based agents are evolving from passive text generators into autonomous systems capable of planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence supported each claim, whether tool calls were justified, how memory influenced later decisions, or where failures originated. This survey examines evidence tracing and execution provenance as foundations for process-level accountability in trustworthy LLM agents. We define execution provenance as the typed graph of an agent execution and evidence tracing as its projection onto evidence-support relations. This perspective connects retrieval grounding, claim support, tool-use safety, memory lineage, observability, debugging, audit, and recovery within a unified framework. We introduce a taxonomy covering trace sources, evidence and execution units, provenance relations, tracing granularity and timing, representation forms, and trust functions. We then review key methodological directions, including provenance representation, evidence attribution, tool-use provenance, runtime guardrails, provenance-bearing memory, observability, and failure diagnosis. Finally, we discuss benchmarks, datasets, metrics, and open challenges for building provenance-aware, auditable, and recoverable agent systems.
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。
