Is GraphRAG superseded?

GraphRAG (Agent / long-term memory): superseded — cited as a baseline and beaten by newer methods. 1 paper(s) critique it, 2 beat it on benchmarks — #16 of 63 most-superseded.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.

Agent / long-term memory

GraphRAG

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

Superseded baseline#16 of 63 most-superseded · first seen Jan 21, 2025

Superseded — cited as a baseline and beaten by newer methods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
0.894 vs 0.748
Recall · [Overall Recall]
Beyond Fact Retrieval: Episodic Memory for RAG with Generative Semantic Workspaces
CAM beats GraphRAG
52.3 vs 45.3
ACC-L · [Single-Document Scenarios (NovelQA)]
CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.