REVIEW 51 cited by
Graph Retrieval-Augmented Generation: A Survey
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Recently, Retrieval-Augmented Generation (RAG) has achieved remarkable success in addressing the challenges of Large Language Models (LLMs) without necessitating retraining. By referencing an external knowledge base, RAG refines LLM outputs, effectively mitigating issues such as ``hallucination'', lack of domain-specific knowledge, and outdated information. However, the complex structure of relationships among different entities in databases presents challenges for RAG systems. In response, GraphRAG leverages structural information across entities to enable more precise and comprehensive retrieval, capturing relational knowledge and facilitating more accurate, context-aware responses. Given the novelty and potential of GraphRAG, a systematic review of current technologies is imperative. This paper provides the first comprehensive overview of GraphRAG methodologies. We formalize the GraphRAG workflow, encompassing Graph-Based Indexing, Graph-Guided Retrieval, and Graph-Enhanced Generation. We then outline the core technologies and training methods at each stage. Additionally, we examine downstream tasks, application domains, evaluation methodologies, and industrial use cases of GraphRAG. Finally, we explore future research directions to inspire further inquiries and advance progress in the field. In order to track recent progress in this field, we set up a repository at \url{https://github.com/pengboci/GraphRAG-Survey}.
Forward citations
Cited by 51 Pith papers
-
BM25 Wins at Scale: A Scaling Study of Retrieval-Augmented Generation Paradigms
On an enterprise corpus scaled from 1.7M to 601M tokens, BM25 beats raw-file agentic search, dense retrieval, and graph RAG at large sizes, crossing near 10M tokens.
-
DocNavRAG: Document-Structured Graph RAG with Stateful Evidence Construction for Complex Document Question Answering
A training-free agentic graph RAG system that navigates document hierarchies and cross-region links, improving answer quality by 7.8% and context sufficiency by 17.7% over the strongest baseline across four CDQA benchmarks.
-
Leveraging Trajectory Graphs for Pre-Execution Error Diagnosis in Agentic LLM Systems
Modeling agent trajectories as action-centric probabilistic graphs lets a GNN warn LLM agents of likely step-level errors before execution, improving pass ratio ~14.7% across four benchmarks.
-
KAMR: Grounding Generation via Knowledge-Aligned Multi-hop Retrieval
Partial-alignment contrastive pretraining plus anchor-then-expand graph retrieval improves multi-hop KG evidence recovery and downstream QA over strong dense and graph RAG baselines.
-
FAIR GraphRAG: A Retrieval-Augmented Generation Approach for Semantic Data Analysis
Embedding FAIR Digital Objects as graph nodes yields a GraphRAG system that measurably improves accuracy, coverage and explainability on biomedical RNA-seq queries versus a non-FAIR baseline.
-
IaC Generation with LLMs: An Error Taxonomy and A Study on Configuration Knowledge Injection
Structured configuration knowledge injected as a knowledge graph significantly improves technical correctness of LLM-generated Terraform code but not alignment with user intent.
-
Youtu-GraphRAG: Vertically Unified Agents for Graph Retrieval-Augmented Complex Reasoning
A schema-bounded agentic framework unifies graph construction and retrieval, reporting SOTA accuracy and up to 90.71% token-cost savings across six benchmarks, plus a new anonymized QA dataset.
-
PG-Agent: An Agent Powered by Page Graph
An MLLM GUI agent that stores past episodes as a page graph and retrieves action guidelines from it improves step success on three benchmarks.
-
DSRAG: A Domain-Specific Retrieval Framework Based on Document-derived Multimodal Knowledge Graph
DSRAG combines a two-layer multimodal knowledge graph with vector search to improve domain-specific question answering, reporting gains over Naive RAG, AutoFlow, and RAGFlow on one 100-question dataset.
-
Graph World Model
The Graph World Model uses action nodes and graph message passing to unify multimodal and graph-structured tasks, but its 'outperforms or matches' claim is contradicted by results on Goodreads.
-
RAG Safety: Exploring Knowledge Poisoning Attacks to Retrieval-Augmented Generation
Inserting a small number of crafted triples into a knowledge graph can flip KG-RAG question answering toward attacker-chosen incorrect answers across four recent systems.
-
KERAG_R: Knowledge-Enhanced Retrieval-Augmented Generation for Recommendation
KERAG_R improves LLM-based top-k recommendation by using a GAT to select relevant KG triples and incorporating them into instruction-tuned prompts, reporting gains over ten baselines on three datasets.
-
BYOKG-RAG: Multi-Strategy Graph Retrieval for Knowledge Graph Question Answering
BYOKG-RAG combines LLM-generated entities, paths, queries, and candidate answers with multiple graph retrieval tools to answer questions over custom knowledge graphs without training data.
-
Weak-to-Strong GraphRAG: Aligning Weak Retrievers with Large Language Models for Graph-based Retrieval Augmented Generation
ReG refines weak graph-retriever supervision with LLM-selected reasoning chains and reorganizes retrieved triples into coherent evidence chains, improving KGQA accuracy, data efficiency, and reasoning token efficiency.
-
EraRAG: Efficient and Incremental Retrieval Augmented Generation for Growing Corpora
EraRAG uses hyperplane-based locality-sensitive hashing to build a hierarchical retrieval graph whose affected regions only are re-summarized when new documents arrive, cutting update cost by up to an order of magnitude.
-
XGraphRAG: Interactive Visual Analysis for Graph-based Retrieval-Augmented Generation
A visual analytics framework helps GraphRAG developers trace suspicious retrievals through graph construction and LLM invocation stages, with a user study suggesting faster, more accurate failure diagnosis than an exi...
-
Respecting Temporal-Causal Consistency: Entity-Event Knowledge Graphs for Retrieval-Augmented Generation
A mention-level entity-event knowledge graph for RAG modestly improves temporal-causal question answering on a new narrative benchmark, with gains mostly coming from adding HyDE-style hypothetical answers.
-
Enhancing Clinical Multiple-Choice Questions Benchmarks with Knowledge Graph Guided Distractor Generation
KGGDG generates harder distractors for medical MCQs by walking a knowledge graph to find misleading paths and feeding them to an LLM, lowering LLM accuracy on most benchmarks tested.
-
Invariant Link Selector for Spatial-Temporal Out-of-Distribution Problem
OOD-Linker selects invariant links in temporal graphs via an information-bottleneck objective and reports a generalization error bound and link-prediction experiments under distribution shift.
-
Climate Finance Bench
Climate Finance Bench releases 330 expert-validated QA pairs on 33 climate reports and shows that retrieval quality, not model capacity, is the main accuracy bottleneck.
-
KnowTrace: Bootstrapping Iterative Retrieval-Augmented Generation with Structured Knowledge Tracing
KnowTrace builds a question-specific knowledge graph during iterative retrieval and uses backtracing to filter useful reasoning steps, improving multi-hop QA and self-bootstrapping.
-
Hypercube-Based Retrieval-Augmented Generation for Scientific Question-Answering
A document hypercube with LLM-extracted labels and hybrid sparse/dense matching improves retrieval and QA accuracy over RAG baselines on three datasets.
-
CRAKEN: Cybersecurity LLM Agent with Knowledge-Based Execution
CRAKEN, an LLM agent combining Self-RAG and Graph-RAG over a CTF writeup database, solves 22% of NYU CTF Bench challenges, three percentage points above the prior D-CIPHER baseline.
-
Toward Reliable Scientific Hypothesis Generation: Evaluating Truthfulness and Hallucination in Large Language Models
A new benchmark (TruthHypo) and a knowledge-grounded hallucination detector (KnowHD) show that grounding scores can partially select truthful LLM-generated biomedical hypotheses, but the result is at risk from knowled...
-
Causal Cartographer: From Mapping to Reasoning Over Counterfactual Worlds
An LLM agent extracts a 975-variable causal graph from 2020 oil-price news and a second agent answers counterfactual queries by step-by-step causal reasoning.
-
KET-RAG: A Cost-Efficient Multi-Granular Indexing Framework for Graph-RAG
KET-RAG uses a PageRank-selected knowledge graph skeleton plus a text-keyword bipartite graph to lower Graph-RAG indexing cost while achieving comparable or better retrieval and generation quality.
-
PIER: Physics-Informed Environmental Retrieval for Time-Series Modeling
PIER augments embedding-based retrieval for lake modeling with a physics-aware stream scored by local verifiers, improving water temperature and dissolved oxygen prediction across 356 lakes.
-
Enhancing Relation Modeling with Social Attributes for Social Media Popularity Prediction
RE-Rag blends semantic and social-attribute relations in retrieval and attention, achieving lower prediction error than prior RAG-based models on large popularity-prediction benchmarks.
-
SCAIR: Schema-Conditioned Agentic Iterative Reasoning for Enterprise Knowledge Graphs
SCAIR, a training-free schema-conditioned agentic KG-RAG method, substantially outperforms existing KG-RAG approaches on a new enterprise CMDB benchmark, but the evaluation has notable confounds.
-
GRACE: Graph-Guided Repository-Aware Code Completion through Hierarchical Code Fusion
GRACE combines a multi-level code graph, hybrid text-structure retrieval, and graph fusion to improve repository-level code completion over vanilla and graph-based RAG baselines.
-
GOSU: Retrieval-Augmented Generation with Global-Level Optimized Semantic Unit-Centric Framework
GOSU globally merges semantic units from text chunks into a unit-centric knowledge graph and uses three-tier keyword retrieval to improve RAG generation quality, according to LLM-judge win rates.
-
T-GRAG: A Dynamic GraphRAG Framework for Resolving Temporal Conflicts and Redundancy in Knowledge Retrieval
A temporal GraphRAG framework that partitions knowledge graphs by timestamp and retrieves at subgraph, node, and knowledge levels outperforms RAG baselines on a new Audi annual-report QA benchmark.
-
MemTool: Optimizing Short-Term Memory Management for Dynamic Tool Calling in LLM Agent Multi-Turn Conversations
MemTool is a short-term memory framework with three modes (autonomous, workflow, hybrid) that lets LLM agents add and remove tools across multi-turn conversations, evaluated over 100 turns on 13+ models.
-
MMGraphRAG: Bridging Vision and Language with Interpretable Multimodal Knowledge Graphs
MMGraphRAG links scene-graph entities from images to text knowledge graph entities via SpecLink, and reports accuracy gains over naive RAG and GraphRAG on multimodal document QA.
-
DyG-RAG: Dynamic Graph Retrieval-Augmented Generation with Event-Centric Reasoning
DyG-RAG builds a dynamic event graph from time-anchored event units and uses timeline retrieval with Time-CoT prompting to answer temporal questions.
-
Graphs Meet AI Agents: Taxonomy, Progress, and Future Opportunities
A survey that groups graph-empowered AI agent research into planning, execution, memory, and multi-agent coordination, plus agents-for-graphs and applications.
-
Agent-based Condition Monitoring Assistance with Multimodal Industrial Database Retrieval Augmented Generation
MindRAG retrieves similar historical vibration recordings and maintenance annotations, then uses LLM agents to generate fault predictions and alarm recommendations for industrial condition monitoring.
-
GPR: Empowering Generation with Graph-Pretrained Retriever
GPR pretrains a two-tower retriever on knowledge graphs using LLM-generated questions from masked triplets and a soft-preference triplet loss, improving KGQA accuracy across datasets and LLMs.
-
UORA: Uniform Orthogonal Reinitialization Adaptation in Parameter-Efficient Fine-Tuning of Large Models
UORA is a LoRA/VeRA-style PEFT method that selectively reinitializes low-magnitude rows and columns of frozen random matrices, reaching LoRA-comparable performance with far fewer trainable parameters.
-
Walk&Retrieve: Simple Yet Effective Zero-shot Retrieval-Augmented Generation via Knowledge Graph Walks
Offline walks over a knowledge graph, verbalized into text and retrieved by embedding similarity, let a single LLM call answer multi-hop questions competitively without any fine-tuning.
-
Aethel: A Reproducible Graph-Retrieval Framework for Multi-Hop Financial Diligence
Graph retrieval improves multi-hop passage recall in small closed pools but does not beat BM25 on a 4,123-chunk open financial corpus.
-
Leveraging Spreading Activation for Improved Document Retrieval in Knowledge-Graph-Based RAG Systems
A spreading-activation retrieval module over an automatically constructed knowledge graph improves multi-hop QA in RAG systems, with up to 39 percentage points gain when combined with chain-of-thought retrieval.
-
Explainable Knowledge Graph Retrieval-Augmented Generation (KG-RAG) with KG-SMILE
KG-SMILE applies perturbation and linear regression to a knowledge graph to attribute which entities and relations drive a GraphRAG system's answers.
-
TRAIL: Joint Inference and Refinement of Knowledge Graphs with Large Language Models
TRAIL couples LLM reasoning with on-the-fly KG insertion and pruning, showing accuracy gains on two of five medical QA benchmarks, with a 3% to 13% improvement claim that overstates the results.
-
Inference Scaled GraphRAG: Improving Multi Hop Question Answering on Knowledge Graphs
Allowing an LLM more inference steps and majority-voted graph actions improves multi-hop knowledge-graph QA scores on GRBench.
-
LlamaRec-LKG-RAG: A Single-Pass, Learnable Knowledge Graph-RAG Framework for LLM-Based Ranking
A KG-enhanced LlamaRec that feeds user-specific relation paths into a Llama-2 ranker reports modest MRR, NDCG, and Recall gains on two benchmarks.
-
DistRAG: Towards Distance-Based Spatial Reasoning in LLMs
Retrieving distance facts from a spatial graph improves LLM answers to direct and nearest-city distance questions, while complex distance-comparison questions remain unsolved.
-
KARE-RAG: Knowledge-Aware Refinement and Enhancement for RAG
Training RAG generators on contrastive knowledge-graph pairs with weighted DPO improves average exact-match by 1.8 to 4.2 points across three model sizes.
-
Optimizing the Interface Between Knowledge Graphs and LLMs for Complex Reasoning
Hyperparameter tuning of Cognee's knowledge graph pipeline yields consistent but uneven gains across three multi-hop QA benchmarks, with best configurations varying by dataset and metric.
-
Multilevel Analysis of Cryptocurrency News using RAG Approach with Fine-Tuned Mistral Large Language Model
A fine-tuned Mistral 7B model produces graph and text summaries, sentiment scores, and stacked meta-summaries of crypto news, but the paper reports no quantitative evaluation.
-
HybridRAG-based LLM Agents for Low-Carbon Optimization in Low-Altitude Economy Networks
HybridRAG merges keyword, vector, and graph retrieval to let an LLM formulate carbon-emission optimization problems for multi-UAV MEC networks, and R2DSAC solves them with a diffusion-regularized SAC plus neuron pruni...
Discussion (0). Continue with ORCID to comment.