source-awareness

This paper introduces ProvenanceGuard, a source-aware verification method for MCP-based LLM agents to address cross-source conflation, where claims are factually correct but attributed to the wrong source.

ProvenanceGuard是一种用于MCP驱动的LLM代理的源感知事实性验证器，它通过分解回答为原子声明、路由到特定源证据、检查支持并验证归因，解决了跨源混淆问题。在医疗领域的评估中，它达到了0.802的块F1和0.858的源准确率。