An implementation of the GraphRAG pipeline (based on the 2024 paper "From Local to Global" by Edge et al.) for query-focused summarization of large text corpora.
university-project course-project knowledge-graph-construction rag hierarchial-language-model document-ingestion qfs query-focused-summarization leiden-algorithm system-2 retrieval-augmented-generation graphrag global-query llm-graph
-
Updated
Nov 5, 2025 - Python