arXiv cs.AI· Saiful Khan, Afrah Farea·· 13 小时前AI 评分38
用业务流程知识图谱为 LLM 智能体提供依据：面向可靠多步服务调用的流程感知知识图谱
Grounding LLM Agents in Business Processes: Process-Aware Knowledge Graphs for Reliable Multi-Step Service Invocation
AI 导读
研究者将 OData 元数据与 BPMN 参考流程整合进带类型的属性图，用稠密嵌入加流程边检索候选端点，在 42 个企业服务上评估，低词汇重叠查询的 Recall@3 提升 5.7 个点（n=74）。在 81 条多步查询上，LLM 智能体任务完成率 38.7%，基于分块的基线为 10.3%。提供完整类型化上下文后，无效调用降至 3%（101 次运行中 3 次），且不再编造代码值。
正文 · 原文
Abstract:Large language model (LLM) agents that call enterprise APIs must select the correct endpoints, follow process order, use valid payload keys, and choose coded values from predefined lists. However, current retrieval-augmented service discovery treats endpoints independently and therefore misses process and schema constraints. We address this limitation by integrating Open Data Protocol (OData) metadata and Business Process Model and Notation (BPMN) reference processes in a typed property graph. A hybrid retriever identifies candidate endpoints using dense embeddings and follows process edges to retrieve subsequent endpoints. We evaluate the method on 42 enterprise services with process-chain ground truth. For queries with low lexical overlap between the query and downstream entities, it improves Recall@3 by 5.7 points (n=74), with consistent gains across embedding models and cut-offs. It matches dense retrieval on other queries and provides no improvement on OpenAPI services without a process layer, indicating that the gains arise from process structure. We also evaluate an LLM agent on 81 multi-step queries. It completes 38.7% of tasks, compared with 10.3% for a chunk-based baseline. In an elicitation probe, providing valid code lists eliminates invented codes but leaves invalid keys in 46% of write runs. Full typed context reduces invalid calls to 3% (3 of 101 runs), with no invented codes. These results show that exposing process and schema constraints during retrieval improves the reliability of agent-driven API calls.
| Subjects: | Artificial Intelligence (cs.AI); Software Engineering (cs.SE) |
| Cite as: | arXiv:2610.37242 [cs.AI] |
| (or arXiv:2610.37242v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.37242 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Saiful Khan Dr [view email]
[v1]
Tue, 27 Oct 2026 13:45:29 UTC (111 KB)
来源：arXiv cs.AI · arxiv.org