arXiv：cs.AI· Xin Li, Rami Bahsoon, Tao Chen·· 6 小时前
智能体工作流为何失败：基于文献与实证数据构建的分类体系
Why Agentic Workflows Fail: A Taxonomy Derived from Literature and Empirical Data
AI 导读
一项研究提出面向来源归因的智能体工作流失败分类体系，结合失败类型词典、工作流阶段定位与智能体能力定位，构建出以工作流阶段为横轴、智能体能力为纵轴的双轴定位矩阵。对比显示，文献中失败多集中于规划、推理、学习、感知与知识机制，GitHub 数据中则集中于执行阶段、执行动作、工具、API 与基础设施，最突出的热点是迭代执行与动作执行。研究认为智能体工作流可靠性应被当作系统问题，而非仅模型输出问题。
正文
Abstract:Agentic workflows combine large language models with planning, memory, tools, execution environments, and iterative feedback. Their failures are difficult to diagnose because the same user-visible symptom may originate from model behavior, workflow orchestration, external tools, infrastructure, or their interaction. This paper presents a source-aware taxonomy of agentic workflow failures and applies it to evidence from academic literature and GitHub. The taxonomy combines source attribution with three empirical dimensions: a concrete failure-type dictionary, workflow-stage localization, and agent-capability localization. Its central structure is a two-axis localization matrix: workflow stages form the horizontal process axis and agent capabilities form the vertical functional axis. This design moves beyond taxonomies that mainly classify final failure manifestations by localizing intermediate process states inside agent operation. Our empirical comparison shows that literature-derived failures emphasize planning, reasoning, learning, perception, and knowledge-related mechanisms, whereas GitHub-derived failures concentrate on execution-stage, actuation-related, tool, API, and infrastructure failures. The dominant hotspot is iterative execution and actuation. These findings indicate that agentic workflow reliability should be treated as a systems problem rather than only a model-output problem. The taxonomy provides a basis for structured evaluation, root cause analysis, failure injection, and engineering triage.
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.17418 [cs.SE] |
| (or arXiv:2610.17418v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17418 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xin Li [view email]
[v1]
Mon, 12 Oct 2026 21:32:14 UTC (313 KB)
来源：arXiv：cs.AI · arxiv.org