LangChain's OpenWiki and evidence graph raise agent accuracy on cross-document questions
AILangChain tested OpenWiki, its open-source wiki compiler, as a memory layer for agents on a 111-document synthetic insurance library. On connected questions, the wiki raised accuracy from 0.10 with keyword search to 0.52 and cut cost per correct answer from about $0.55 to $0.19.
Loading source posts…
Source details
LangChain Blog · langchain.comPublished · added here
Original title: Building Evidence-Grounded Agent Memory with OpenWiki and Graphs