KAG, or Knowledge Augmented Generation, is a logical reasoning and Q&A framework that utilizes the OpenSPG engine and large language models to build logical reasoning and Q&A solutions for vertical domain knowledge bases. It effectively overcomes the ambiguity of traditional RAG vector similarity calculations and the noise problem of GraphRAG introduced by OpenIE. KAG supports logical reasoning, multi-hop fact Q&A, and is superior to current methods. Its core features include knowledge and chunk mutual indexing, knowledge alignment using conceptual semantic reasoning, and schema-constrained knowledge construction. Interestingly, KAG proposes a hybrid solution and inference engine guided by logic forms, transforming natural language problems into problem-solving processes. The framework aims to build a knowledge-enhanced LLM service framework in professional domains. ⭐️
https://github.com/OpenSPG/KAG