CoChat
Sign Up
Cited by 389Topic ModelingInformation Retrieval and Search BehaviorMultimodal Machine Learning Applications

LightRAG: Simple and Fast Retrieval-Augmented Generation

Open access
Guo, Zirui, Xia, Lianghao, Yu, Yanhua, Ao, Tu, Huang, Chao·2025·Findings of the Association for Computational Linguistics: EMNLP 2025
DOI 10.18653/v1/2025.findings-emnlp.568
Open this paper in CoChat — free accountOpen PDF

Abstract

Retrieval-Augmented Generation (RAG) systems enhance large language models (LLMs) by integrating external knowledge sources, enabling more accurate and contextually relevant responses tailored to user needs.However, existing RAG systems have significant limitations, including reliance on flat data representations and inadequate contextual awareness, which can lead to fragmented answers that fail to capture complex interdependencies.To address these challenges, we propose Ligh-tRAG, a novel framework that incorporates graph structures into text indexing and retrieval processes.This innovative approach employs a dual-level retrieval system that enhances comprehensive information retrieval from both lowand high-level knowledge discovery.Additionally, the integration of graph structures with vector representations facilitates efficient retrieval of related entities and their relationships, significantly improving response times while maintaining contextual relevance.This capability is further enhanced by an incremental update algorithm that ensures the timely integration of new data, allowing the system to remain effective and responsive in rapidly changing data environments.Extensive experimental validation demonstrates considerable improvements in retrieval accuracy and efficiency compared to existing approaches.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.18653/v1/2025.findings-emnlp.568

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Current research on LLM-generated knowledge graphs has primarily operated without formal ontologies.

    “Additionally, while current research on LLM-generated knowledge graphs has primarily operated without formal ontologies [9][4][17][19][26], domain-specific ontologies present significant potential.”

    From Zep: A Temporal Knowledge Graph Architecture for Agent Memory (2025) · cites this paper as [17]

    Section 3.1 describes LightRAG using an LLM to extract entities and relationships for a knowledge graph, and Section 4.1 describes prompts using specified entity types; neither describes a formal ontology. The paper therefore provides an example consistent with the claim, but it does not establish that current research has primarily operated without formal ontologies. (LLM, conf=0.78)

    Experimental — accuracy not yet measured

  • Knowledge graph-based methods enhance semantic connectivity through structured representations, but their partial graph construction prevents establishing relationships between temporally distant nodes across conversation sessions.

    “Knowledge graph-based methods (Edge et al., 2025; Guo et al., 2025; Zhu et al., 2025) enhance semantic connectivity through structured representations, but their partial graph construction prevents establishing relationships between temporally distant nodes across conversation sessions.”

    From Pre-Storage Reasoning for Episodic Memory: Shifting Inference Burden to Memory for Personalized Dialogue (2025) · cites this paper as Guo et al., 2025

    The Introduction (page unavailable) says graph structures represent interdependencies among entities and help synthesize information from multiple sources, supporting the semantic-connectivity claim. Section 3.1 (page unavailable) describes constructing a knowledge graph across a document collection, but the paper does not say that partial graph construction prevents relationships between temporally distant nodes across conversation sessions. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Zep's community search approach parallels the high-level key search methodology in LightRAG.

    “While developed independently, our community search approach parallels the high-level key search methodology in LightRAG [17].”

    From Zep: A Temporal Knowledge Graph Architecture for Agent Memory (2025) · cites this paper as [17]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Still being checked.

Research works printed but not linked: 21

  1. William Brannon, Suyash Fulay, Hang Jiang, Won- june Kang, Brandon Roy, Jad Kabbara, and Deb Roy. 2023. Congrat: Self-supervised contrastive pre- training for joint graph and text embeddings. arXiv preprint arXiv:2305.14321. · arXiv
  2. Chi-Min Chan, Chunpu Xu, Ruibin Yuan, Hongyin Luo, Wei Xue, Yike Guo, and Jie Fu. 2024. Rq-rag: Learn- ing to refine queries for retrieval augmented genera- tion. arXiv preprint arXiv:2404.00610. · arXiv
  3. Runjin Chen, Tong Zhao, AJAY KUMAR JAISWAL, Neil Shah, and Zhangyang Wang. 2024. Llaga: Large language and graph assistant. In ICML.
  4. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, and Jonathan Larson. 2024. From local to global: A graph rag approach to query-focused summarization. arXiv preprint arXiv:2404.16130. · arXiv
  5. Shahul Es, Jithin James, Luis Espinosa Anke, and Steven Schockaert. 2024. Ragas: Automated evalua- tion of retrieval augmented generation. In EACL.
  6. Wenqi Fan, Yujuan Ding, Liangbo Ning, Shijie Wang, Hengyun Li, Dawei Yin, Tat-Seng Chua, and Qing Li. 2024. A survey on rag meeting llms: Towards retrieval-augmented large language models. In KDD.
  7. Luyu Gao, Xueguang Ma, Jimmy Lin, and Jamie Callan. 2022. Precise zero-shot dense retrieval without rele- vance labels. arXiv preprint arXiv:2212.10496. · arXiv
  8. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, and Haofen Wang. 2023. Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:2312.10997. · arXiv
  9. Yichuan Li, Kaize Ding, and Kyumin Lee. 2023. Grenade: Graph-centric language model for self- supervised representation learning on text-attributed graphs. In EMNLP.
  10. Hao Liu, Jiarui Feng, Lecheng Kong, Ningyue Liang, Dacheng Tao, Yixin Chen, and Muhan Zhang. 2024. One for all: Towards training one graph model for all classification tasks. In ICLR.
  11. Yuanjie Lyu, Zhiyu Li, Simin Niu, Feiyu Xiong, Bo Tang, Wenjin Wang, Hao Wu, Huanyong Liu, Tong Xu, and Enhong Chen. 2024. Crud-rag: A comprehensive chinese benchmark for retrieval- augmented generation of large language models. arXiv preprint arXiv:2401.17043. · arXiv
  12. Hongjin Qian, Peitian Zhang, Zheng Liu, Kelong Mao, and Zhicheng Dou. 2024. Memorag: Moving to- wards next-gen rag via memory-inspired knowledge discovery. arXiv preprint arXiv:2409.05591. · arXiv
  13. Ori Ram, Yoav Levine, Itay Dalmedigos, Dor Muhlgay, Amnon Shashua, Kevin Leyton-Brown, and Yoav Shoham. 2023. In-context retrieval-augmented lan- guage models. TACL, 11:1316-1331.
  14. Ladislav Rampášek, Michael Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini. 2022. Recipe for a general, powerful, scal- able graph transformer. NeurIPS, 35:14501-14515.
  15. Alireza Salemi et al. 2024. Evaluating retrieval quality in retrieval-augmented generation. In SIGIR.
  16. Viju Sudhi, Sinchana Ramakanth Bhat, Max Rudat, et al. 2024. Rag-ex: A generic framework for explaining retrieval augmented generation. In SIGIR.
  17. Jiabin Tang, Yuhao Yang, Wei Wei, Lei Shi, Lixin Su, Suqi Cheng, Dawei Yin, and Chao Huang. 2024. Graphgpt: Graph instruction tuning for large lan- guage models. In SIGIR.
  18. Shangqing Tu, Yuanchun Wang, Jifan Yu, Yuyang Xie, Yaran Shi, Xiaozhi Wang, Jing Zhang, Lei Hou, and Juanzi Li. 2024. R-eval: A unified toolkit for evaluat- ing domain knowledge of retrieval augmented large language models. In KDD.
  19. Han Xie, Da Zheng, Jun Ma, Houyu Zhang, Vassilis N Ioannidis, Xiang Song, Qing Ping, et al. 2023. Graph- aware language model pre-training on a large graph corpus can help multiple graph applications. In KDD.
  20. Yue Yu, Wei Ping, Zihan Liu, Boxin Wang, Jiaxuan You, Chao Zhang, Mohammad Shoeybi, and Bryan Catanzaro. 2024. Rankrag: Unifying context ranking with retrieval-augmented generation in llms. arXiv preprint arXiv:2407.02485. · arXiv
  21. Penghao Zhao, Hailin Zhang, Qinhan Yu, Zhengren Wang, Yunteng Geng, Fangcheng Fu, Ling Yang, Wentao Zhang, and Bin Cui. 2024. Retrieval- augmented generation for ai-generated content: A survey. arXiv preprint arXiv:2402.19473. · arXiv

Cited by

Sorted by citations

Related

Authors
Guo, Z., Xia, L., Yu, Y., Ao, T., & Huang, C.
Affiliations
  • Beijing University of Posts and Telecommunications (CN)
  • University of Hong Kong (HK)
ORCID
Venue
Findings of the Association for Computational Linguistics: EMNLP 2025
Publisher
Association for Computational Linguistics
Published
2025
Pages
10746-10761
DOI
10.18653/v1/2025.findings-emnlp.568
OpenAlex
W4416034797
Cited by
208
References
0
Citations by year
2026: 159 · 2025: 49
FWCI
198.29
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
  • Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion, cc-by-nc-sa)
  • arXiv (Cornell University)
Funders
National Natural Science Foundation of China (U22B2019)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
Keywords
retrieval-augmented generation, vector representation, large language models, external knowledge integration, context awareness

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review