Is GraphRAG superseded?

GraphRAG (Agent / long-term memory): superseded — cited as a baseline and beaten by newer methods. 1 paper(s) critique it, 2 beat it on benchmarks — #16 of 63 most-superseded. Sub-problem: cluster led by RAPTOR. Newer alternatives in the same sub-problem include HingeMem, AtomMem, REMem, Generative Semantic Workspace (GSW), CAM.

Method Drift›Agent / long-term memory

Superseded baseline#16 of 63 most-superseded

GraphRAG

A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models

Agent / long-term memory · first seen Jan 21, 2025

superseded — cited as a baseline and beaten by newer methods

1 papers critique it · 2 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites GraphRAG as a baseline.

“Although these works effectively structure large-scale textual data to enhance retrieval and generation capabilities, they are confined to static corpora, requiring complete reconstruction to integrate new information.”
— CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension

Beaten on benchmarks

Head-to-head results where a newer method reports beating GraphRAG. Values are copied from the source paper's tables — verify against the cited paper.

GSW beats GraphRAG · Precision [Overall Precision]
0.865 vs 0.781
Beyond Fact Retrieval: Episodic Memory for RAG with Generative Semantic Workspaces
GSW beats GraphRAG · Recall [Overall Recall]
0.894 vs 0.748
Beyond Fact Retrieval: Episodic Memory for RAG with Generative Semantic Workspaces
GSW beats GraphRAG · F1 [Overall F1]
0.850 vs 0.714
Beyond Fact Retrieval: Episodic Memory for RAG with Generative Semantic Workspaces
CAM beats GraphRAG · R-L [Single-Document Scenarios (NovelQA)]
25.4 vs 22.5
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · ACC-L [Single-Document Scenarios (NovelQA)]
52.3 vs 45.3
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · R-L [Single-Document Scenarios (QMSum)]
26.5 vs 25.2
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · ACC-L [Single-Document Scenarios (QMSum)]
57.6 vs 53.9
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · F1P [Single-Document Scenarios (FABLES)]
91.5 vs 87.2
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · F1N [Single-Document Scenarios (FABLES)]
52.5 vs 48.0
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · EM [Multi-Document Scenarios (MH-RAG)]
72.8 vs 65.2
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · F1 [Multi-Document Scenarios (MH-RAG)]
77.5 vs 70.3
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension
CAM beats GraphRAG · R-L [Multi-Document Scenarios (ODSum-Story)]
25.5 vs 24.3
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.