AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

HGMem: Hypergraph-based Working Memory to Improve Multi-step RAG for Long-Context Complex Relational Modeling

Chulun Zhou, Chunkang Zhang, Guoxin Yu, Fandong Meng, Jie Zhou, Wai Lam, Mo Yu

ICML 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 0 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 merging operation to memory evolution (ablation)HGMem (Qwen2.5-32B) · NarrativeQAaccuracy: 43.00 -> 51.00Improved
add merging operation to memory evolution (ablation)HGMem (Qwen2.5-32B) · Longbenchdiversity: 61.80 -> 66.51Improved
add merging operation to memory evolution (ablation)HGMem (Qwen2.5-32B) · Preludeaccuracy: 57.78 -> 62.22Improved
add update operation to memory evolution (ablation)HGMem (Qwen2.5-32B) · Longbenchcomprehensiveness: 62.48 -> 64.18Improved
add update operation to memory evolution (ablation)HGMem (Qwen2.5-32B) · NarrativeQAaccuracy: 50.00 -> 51.00Improved
use adaptive retrieval instead of Global Exploration only (ablation)HGMem (Qwen2.5-32B) · Longbenchcomprehensiveness: 59.25 -> 64.18Improved
use adaptive retrieval instead of Local Investigation only (ablation)HGMem (Qwen2.5-32B) · NoChaaccuracy: 63.49 -> 70.63Improved
add merging on sense-making queries (ablation)HGMem (Qwen2.5-32B) · NarrativeQA sense-making queriesaccuracy: 30.00 -> 40.00Improved
add merging on sense-making queries (ablation)HGMem (Qwen2.5-32B) · NarrativeQA sense-making queriesavg entities per hyperedge: 4.10 -> 7.07Improved
add merging on primitive queries (ablation)HGMem (Qwen2.5-32B) · NoCha primitive queriesaccuracy: 65.00 -> 60.00Hurt
replace offline graph with 50% ablated graphHGMem (Qwen2.5-32B) · Preludeaccuracy: 62.22 -> 57.97Hurt
replace LLM-built graph with Stanford OpenIE graphHGMem (Qwen2.5-32B) · NoChaaccuracy: 70.63 -> 66.67Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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