用知识图谱精准路由工具，让AI问答更准更快还带出处。
SchemaRouter: Field-Aware Tool Routing for Efficient Heterogeneous Agentic RAG
- 构建工具与字段的语义图谱，通过意图匹配自动选工具和字段。
- 准确率0.71，只用227个上下文词元，延迟比基线低2.7倍。
- 93%工具调用正确，62%答案带来源和授权信息，适合科研场景。
异构智能体检索增强生成系统日益集成外部API、内部数据库、向量库和图数据库。直接暴露所有工具描述或仅靠向量相似性选择工具，会导致过度获取（增加负载、耗时）和不足获取（遗漏关键字段）。我们提出SchemaRouter，一个轻量级路由层，将工具、端点、参数、响应字段、领域概念、单位、来源和许可证政策表示为模式图。给定查询后，SchemaRouter生成可执行的工具计划，指定调用哪些工具及获取哪些字段。小模型提取意图、概念和源约束，字段选择通过意图分组投影与概念-字段匹配（含别名层）实现。在包含110个查询的材料科学基准上，SchemaRouter准确率达0.71，在重叠置信区间内媲美“全量获取”方法，超过“提示全部”的0.66（二者置信区间重叠）。其仅使用227个检索上下文词元，远低于全量获取的2,066个；端到端延迟比“提示全部”降低2.7倍。同时获得最佳工具精确率0.93和参数有效性1.0。62%的答案包含来源和许可证信息，而所有基线约为0%。我们发现最小化字段数量会适得其反：准确率降至0.56，仅节省少量词元；而保留召回率的投影策略可恢复最高准确率。SchemaRouter在保持竞争力准确率的同时，提升了效率、与模式规模无关的可扩展性，以及可验证的来源与授权信息支持。
原文摘要 · Abstract (English)
Heterogeneous agentic retrieval-augmented generation (RAG) systems increasingly orchestrate external APIs, internal databases, vector stores, and graph stores. Exposing all tool descriptions to an LLM agent, or selecting tools only by vector similarity, causes two costly failures: over-fetching, which increases payload size, token use, and latency, and under-fetching, which omits fields needed to answer the query. We present SchemaRouter, a lightweight routing layer that represents tools, endpoints, parameters, response fields, domain concepts, units, provenance, and license policies as a schema graph. Given a query, SchemaRouter emits an executable tool plan specifying which tools to call and which fields to retrieve. A small LLM extracts intent, concepts, and source constraints, while field selection is deterministic over the graph through intent-group projection and concept-field matching with an alias layer. On a materials-science benchmark of 110 queries, SchemaRouter achieves answer accuracy of 0.71, matching fetch-everything within overlapping confidence intervals and exceeding prompt-all's 0.66, though their intervals overlap. It uses 227 retrieved-context tokens versus 2,066 for fetch-everything and achieves 2.7x lower end-to-end latency than prompt-all. It also obtains the best tool-exact rate of 0.93 and parameter validity of 1.0. SchemaRouter grounds provenance and license information in 62 percent of answers, compared with approximately 0 percent for all baselines. We also find that minimizing selected-field count is counterproductive: it reduces answer accuracy to 0.56 with negligible token savings, while recall-preserving projection restores top accuracy. SchemaRouter improves efficiency, schema-size-independent scaling, and verifiable provenance/license-grounded answering at competitive accuracy.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。