AI Paper Atlas

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

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 51 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace Leiden communities in GraphRAG with attributed communities (ablation)GraphRAG global · HotpotQAaccuracy (%): 33.5 -> 50.6Improved
replace Leiden communities in GraphRAG with attributed communities (ablation)GraphRAG global · Multihop-RAGaccuracy (%): 45.9 -> 49.3Improved
add hierarchical multi-layer communities (vs single layer) (ablation)ArchRAG · Multihop-RAGaccuracy (%): 63.8 -> 68.8Improved
add hierarchical multi-layer communities (vs single layer) (ablation)ArchRAG · HotpotQAaccuracy (%): 60.1 -> 65.4Improved
add communities in addition to entities (ablation)ArchRAG · HotpotQAaccuracy (%): 59.9 -> 65.4Improved
use LLM-based clustering framework instead of plain Leiden (ablation)ArchRAG · Multihop-RAGaccuracy (%): 63.2 -> 68.8Improved
use weighted Leiden instead of spectral clustering (ablation)ArchRAG · HotpotQAaccuracy (%): 60.5 -> 65.4Improved
use graph augmentation with spectral clustering (ablation)ArchRAG variant · HotpotQAaccuracy (%): 64.8 (no aug) vs 60.5 (with aug)Mixed
add adaptive filtering-based generation (ablation)ArchRAG · HotpotQAaccuracy (%): 40.7 -> 65.4Improved
add adaptive filtering-based generation (ablation)ArchRAG · Multihop-RAGrecall (%): 29.6 -> 37.2Improved
use ArchRAG instead of GraphRAG global search for token costArchRAG · HotpotQAtoken usage (M): 1,394M -> 5.1MImproved
use ArchRAG instead of GraphRAG global on abstract QA diversity and empowermentArchRAG · Multihop-RAG abstract QALLM win rateNo 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.

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy