ContextRAG: Extraction-Free Hierarchical Graph Construction for Retrieval-Augmented Generation

ArXi:2605.19735v1 Announce Type: cross Graph-structured retrieval-augmented generation (RAG) systems can improve answer quality on multi-hop questions, but many current systems rely on large language models (LLMs) to extract entities, relations, and summaries during indexing. These calls add token and wall-clock costs that grow with corpus size. We present ContextRAG, a graph RAG system whose graph topology is constructed without LLM-based entity or relation extraction.