E2065 - RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM

🤗 Upvotes: 115 | cs.CL, cs.AI

Authors:
Mikhail Komarov, Ivan Bondarenko, Stanislav Shtuka, Oleg Sedukhin, Roman Shuvalov, Yana Dementyeva, Matvey Solovyov, Nikolay O. Nikitin

Title:
RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM

Arxiv:
http://arxiv.org/abs/2607.11683v1

Abstract:
Graph retrieval-augmented generation (GraphRAG) enhances large language models with structured knowledge, yet existing systems construct knowledge graphs in a single extraction pass, producing noisy entities and brittle retrieval. RAGU, an open-source modular GraphRAG engine, addresses this by separating extraction from consolidation: entities and relations pass through two...