{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T09:03:03Z","timestamp":1779872583354,"version":"3.53.1"},"reference-count":31,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T00:00:00Z","timestamp":1779840000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Applied Basic Research Project of Liaoning Province","award":["2022JH2\/101300270"],"award-info":[{"award-number":["2022JH2\/101300270"]}]},{"name":"Promote Scientific Research Cooperation with Canada, Australia, New Zealand and Latin America and High-level Talent Training Program of Ministry of Education of China","award":["Liujin Mei2023 No.25"],"award-info":[{"award-number":["Liujin Mei2023 No.25"]}]},{"name":"General Program of the National Social Science Fund of China","award":["24BTQ033"],"award-info":[{"award-number":["24BTQ033"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>The escalating volume and complexity of scientific publications have intensified the demand for intelligent literature survey tools. Current systems, whether based on large language models (LLMs) or retrieval augmented generation (RAG), often struggle with terminological accuracy and semantic coherence, limiting their practical adoption. To address this, we propose a novel LLM-driven framework that synergizes multi-agent collaboration with RAG to enhance both the structural coherence of full-document surveys and the factual reliability of generated content. On the SciReviewGen benchmark, our model achieves Recall-Oriented Understudy for Gisting Evaluation (ROUGE)-1\/2\/L scores of 42.84\/16.57\/18.25, outperforming strong baselines including Query-weighted Fusion-in-Decoder (QFiD), Fusion-in-Decoder (FiD), BigBird, GPT-4o-mini, and Llama4-17B-16E. The framework also excels in human-aligned evaluation, attaining an LLM-as-judge score of 0.70, surpassing QFiD by +0.29. Ablation studies confirm the critical role of both RAG and multi-agent design: removing RAG reduces ROUGE-2\/L by 2.28\/3.19, while single-agent ablation further decreases ROUGE-1 by 1.15. Our work not only advances the state of the art in automated literature synthesis but also offers a scalable solution to mounting scholarly information overload.<\/jats:p>","DOI":"10.7717\/peerj-cs.3882","type":"journal-article","created":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T08:18:22Z","timestamp":1779869902000},"page":"e3882","source":"Crossref","is-referenced-by-count":0,"title":["A large language model-driven scientific literature surveys generation framework based on multi-agents and retrieval-augmented generation"],"prefix":"10.7717","volume":"12","author":[{"given":"Ruihua","family":"Qi","sequence":"first","affiliation":[{"name":"School of Software, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"},{"name":"Research Center of Language Intelligence, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-5841-7105","authenticated-orcid":true,"given":"Haobo","family":"Lyu","sequence":"additional","affiliation":[{"name":"School of Software, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weilong","family":"Li","sequence":"additional","affiliation":[{"name":"School of Software, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuqin","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Software, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xu","family":"Guo","sequence":"additional","affiliation":[{"name":"School of Software, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"},{"name":"Research Center of Language Intelligence, Dalian University of Foreign Languages, Dalian, Liaoning Province, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,5,27]]},"reference":[{"key":"10.7717\/peerj-cs.3882\/ref-1","first-page":"8","article-title":"Towards multi-document summarization of scientific articles: making interesting comparisons with SciSumm","author":"Agarwal","year":"2011"},{"key":"10.7717\/peerj-cs.3882\/ref-2","first-page":"8","article-title":"Using LLM (large language model) to improve efficiency in literature review for undergraduate research","author":"Antu","year":"2023"},{"issue":"1","key":"10.7717\/peerj-cs.3882\/ref-3","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1186\/s13643-018-0740-7","article-title":"Making progress with the automation of systematic reviews: principles of the international collaboration for the automation of systematic reviews (ICASR)","volume":"7","author":"Beller","year":"2018","journal-title":"Systematic Reviews"},{"key":"10.7717\/peerj-cs.3882\/ref-4","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1903.10676","article-title":"SciBERT: a pretrained language model for scientific text","author":"Beltagy","year":"2019"},{"key":"10.7717\/peerj-cs.3882\/ref-5","doi-asserted-by":"publisher","first-page":"115119\u2013115145","DOI":"10.52202\/079017-3655","article-title":"AutoSurvey: large language models can automatically write surveys","volume":"37","author":"Dai","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.7717\/peerj-cs.3882\/ref-6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2404.16130","article-title":"From local to global: a graph rag approach to query-focused summarization","author":"Edge","year":"2024"},{"key":"10.7717\/peerj-cs.3882\/ref-7","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1613\/jair.1523","article-title":"LexRank: graph-based lexical centrality as salience in text summarization","volume":"22","author":"Erkan","year":"2004","journal-title":"Journal of Artificial Intelligence Research"},{"key":"10.7717\/peerj-cs.3882\/ref-8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2007.01282","article-title":"Leveraging passage retrieval with generative models for open domain question answering","author":"Izacard","year":"2020"},{"issue":"12","key":"10.7717\/peerj-cs.3882\/ref-9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3571730","article-title":"Survey of hallucination in natural language generation","volume":"55","author":"Ji","year":"2023","journal-title":"ACM Computing Surveys"},{"key":"10.7717\/peerj-cs.3882\/ref-10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2307.10169","article-title":"Challenges and applications of large language models","author":"Kaddour","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-11","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.15186","article-title":"SciReviewGen: a large-scale dataset for automatic literature review generation","author":"Kasanishi","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-12","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1007\/978-981-97-9443-0_43","article-title":"Instruct large language models to generate scientific literature survey step by step","volume-title":"Natural Language Processing and Chinese Computing","author":"Lai","year":"2025"},{"key":"10.7717\/peerj-cs.3882\/ref-13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2510.24476","article-title":"Mitigating hallucination in large language models (LLMs): an application-oriented survey on rag, reasoning, and agentic systems","author":"Li","year":"2025"},{"key":"10.7717\/peerj-cs.3882\/ref-14","first-page":"418","article-title":"Cluster-based effective generation of AI-driven literature surveys","author":"Li","year":"2024"},{"key":"10.7717\/peerj-cs.3882\/ref-15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2307.03172","article-title":"Lost in the middle: how language models use long contexts","author":"Liu","year":"2023"},{"issue":"1","key":"10.7717\/peerj-cs.3882\/ref-16","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1186\/s13643-019-1074-9","article-title":"Toward systematic review automation: a practical guide to using machine learning tools in research synthesis","volume":"8","author":"Marshall","year":"2019","journal-title":"Systematic Reviews"},{"key":"10.7717\/peerj-cs.3882\/ref-17","first-page":"584","article-title":"Using citations to generate surveys of scientific paradigms","author":"Mohammad","year":"2009"},{"key":"10.7717\/peerj-cs.3882\/ref-18","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1808.08745","article-title":"Don\u2019t give me the details, just the summary. Topic-Aware Convolutional Neural Networks for Extreme Summarization","author":"Narayan","year":"2018"},{"key":"10.7717\/peerj-cs.3882\/ref-19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2505.20096","article-title":"MA-RAG: multi-agent retrieval-augmented generation via collaborative chain-of-thought reasoning","author":"Nguyen","year":"2025"},{"issue":"5","key":"10.7717\/peerj-cs.3882\/ref-20","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3234150","article-title":"A survey on deep learning: algorithms, techniques, and applications","volume":"51","author":"Pouyanfar","year":"2018","journal-title":"ACM Computing Surveys (CSUR)"},{"key":"10.7717\/peerj-cs.3882\/ref-21","first-page":"169","article-title":"Generation of scientific literature surveys based on large language models (LLM) and multi-agent systems (MAS)","author":"Qi","year":"2024"},{"key":"10.7717\/peerj-cs.3882\/ref-22","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2023.emnlp-main.155","article-title":"The troubling emergence of hallucination in large language models\u2014an extensive definition, quantification, and prescriptive remediations","author":"Rawte","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2302.13971","article-title":"LLaMA: open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-24","doi-asserted-by":"publisher","first-page":"106589","DOI":"10.1016\/j.infsof.2021.106589","article-title":"Automation of systematic literature reviews: a systematic literature review","volume":"136","author":"Van Dinter","year":"2021","journal-title":"Information and Software Technology"},{"key":"10.7717\/peerj-cs.3882\/ref-25","doi-asserted-by":"publisher","first-page":"5998","DOI":"10.65215\/ctdc8e75","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"6","key":"10.7717\/peerj-cs.3882\/ref-26","doi-asserted-by":"publisher","first-page":"299","DOI":"10.3390\/info15060299","article-title":"The era of artificial intelligence deception: unraveling the complexities of false realities and emerging threats of misinformation","volume":"15","author":"Williamson","year":"2024","journal-title":"Information"},{"key":"10.7717\/peerj-cs.3882\/ref-27","first-page":"28489","article-title":"Agentic reasoning: a streamlined framework for enhancing LLM reasoning with agentic tools","author":"Wu","year":"2025"},{"key":"10.7717\/peerj-cs.3882\/ref-28","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2007.14062","article-title":"Big bird: transformers for longer sequences","author":"Zaheer","year":"2020"},{"key":"10.7717\/peerj-cs.3882\/ref-29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2309.01219","article-title":"Siren\u2019s song in the Ai ocean: a survey on hallucination in large language models","author":"Zhang","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-30","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.18223","article-title":"A survey of large language models","author":"Zhao","year":"2023"},{"key":"10.7717\/peerj-cs.3882\/ref-31","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2306.05685","article-title":"Judging LLM-as-a-judge with MT-bench and chatbot arena","author":"Zheng","year":"2023"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3882.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3882.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3882.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3882.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T08:18:31Z","timestamp":1779869911000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3882"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,27]]},"references-count":31,"alternative-id":["10.7717\/peerj-cs.3882"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3882","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,27]]},"article-number":"e3882"}}