One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG
For enterprise RAG systems, this work addresses the session-level information need gap, showing that session coverage is a more relevant metric than single-query recall.
RAG systems retrieve documents per query, but enterprise users have sessions of related questions. The proposed method reorganizes the knowledge base via co-occurrence-aware clustering and expands retrieval via cluster neighborhoods, raising session coverage from 41% to 58% and reducing retrieval calls by 34% while compressing the KB to 20% of its original size.
RAG systems retrieve documents optimized for answering one query at a time. Yet enterprise users arrive with sessions, that is, coherent episodes of related questions that span semantically distant parts of the knowledge base. We show that a single retrieval call over a standard knowledge base covers only 41% of a user's session-level information need. To close this gap, we reorganize the KB offline using co-occurrence-aware clustering and expand retrieval candidates through cluster neighborhoods at query time. On WixQA (6,221 enterprise support articles), our method raises single-query session coverage to 58% (+17% absolute; 95% CI: [14.1, 20.4]), reduces retrieval calls to 70% coverage by 34%, and compresses the KB to 20% of its original size, all consistently across four embedding models and six functional domains. We argue that session-level coverage, not single-query recall, should be the primary metric for enterprise RAG evaluation.