Research map / Large language models / Retrieval and knowledge
ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation
Shu Wang, Yixiang Fang, Yingli Zhou, Xilin Liu, Yuchi Ma
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace Leiden communities in GraphRAG with attributed communities (ablation) | GraphRAG global · HotpotQA | accuracy (%): 33.5 -> 50.6 | Improved |
| replace Leiden communities in GraphRAG with attributed communities (ablation) | GraphRAG global · Multihop-RAG | accuracy (%): 45.9 -> 49.3 | Improved |
| add hierarchical multi-layer communities (vs single layer) (ablation) | ArchRAG · Multihop-RAG | accuracy (%): 63.8 -> 68.8 | Improved |
| add hierarchical multi-layer communities (vs single layer) (ablation) | ArchRAG · HotpotQA | accuracy (%): 60.1 -> 65.4 | Improved |
| add communities in addition to entities (ablation) | ArchRAG · HotpotQA | accuracy (%): 59.9 -> 65.4 | Improved |
| use LLM-based clustering framework instead of plain Leiden (ablation) | ArchRAG · Multihop-RAG | accuracy (%): 63.2 -> 68.8 | Improved |
| use weighted Leiden instead of spectral clustering (ablation) | ArchRAG · HotpotQA | accuracy (%): 60.5 -> 65.4 | Improved |
| use graph augmentation with spectral clustering (ablation) | ArchRAG variant · HotpotQA | accuracy (%): 64.8 (no aug) vs 60.5 (with aug) | Mixed |
| add adaptive filtering-based generation (ablation) | ArchRAG · HotpotQA | accuracy (%): 40.7 -> 65.4 | Improved |
| add adaptive filtering-based generation (ablation) | ArchRAG · Multihop-RAG | recall (%): 29.6 -> 37.2 | Improved |
| use ArchRAG instead of GraphRAG global search for token cost | ArchRAG · HotpotQA | token usage (M): 1,394M -> 5.1M | Improved |
| use ArchRAG instead of GraphRAG global on abstract QA diversity and empowerment | ArchRAG · Multihop-RAG abstract QA | LLM win rate | No clear effect |
Later papers that compared with this method
1 later accepted paper reports ArchRAG as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- T-Retriever: Tree-based Hierarchical Retrieval Augmented Generation for Textual Graphs (AAAI 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- RAPTOR — RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval (ICLR 2024) · better on 3 of 3 shared results
- HippoRAG — HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (NeurIPS 2024) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layers (NeurIPS 2025 · similarity 0.90)
- GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (NeurIPS 2025 · similarity 0.89)
- HyperGraphRAG: Retrieval-Augmented Generation via Hypergraph-Structured Knowledge Representation (NeurIPS 2025 · similarity 0.88)
- Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need? (NeurIPS 2025 · similarity 0.88)
- SiReRAG: Indexing Similar and Related Information for Multihop Reasoning (ICLR 2025 · similarity 0.88)
- Fast Think-on-Graph: Wider, Deeper and Faster Reasoning of Large Language Model on Knowledge Graph (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
