AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Iterative Multi-Granular RAG with Contextual Hierarchical Graph

Yanli Hu, Teng Liu, Zhuangyi Zhou, Weixin Zeng, Zhen Tan, Xiang Zhao

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 2 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
add Contextual Hierarchical Graph (ablation)Naive RAG · 2Wiki and HotpotQARecall@5: 2Wiki 61.5 -> 80.1; HotpotQA 75.3 -> 97.0Improved
add Contextual Hierarchical Graph (ablation)Naive RAG · PopQARecall@5: 56.6 -> 56.6No clear effect
add query decomposition (ablation)Naive RAG + CHG · PopQA, 2Wiki, HotpotQARecall@5: PopQA 56.6 -> 58.7; 2Wiki 80.1 -> 84.5; HotpotQA 97.0 -> 97.4Improved
add evidence-augmented reranking (ablation)MGranRAG · PopQA, 2Wiki, HotpotQARecall@5: PopQA 58.7 -> 72.9; 2Wiki 84.5 -> 93.7; HotpotQA 97.4 -> 98.2Improved
increase iterations from T=1 to T=3 (ablation)MGranRAG (Qwen3-8B) · average of 6 benchmarksRecall@5: 83.3 -> 84.2Improved
increase candidate passage pool M (ablation)MGranRAG · 2Wiki (T=1)Recall@5: around 96% at M=10/15Improved
use NV-Embed-v2 instead of Contriever retriever (ablation)MGranRAG · 2WikiRecall@5Improved
use MGranRAG instead of HippoRAG 2 for QAMGranRAG · average F1 6 benchmarksF1: 55.0 -> 56.2Improved
increase iterations T=1 to T=3 (ablation)MGranRAG · LLM token cost on 2Wikiretrieval input tokens: 4.5M -> 12.0MHurt

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