{"doi":"10.1093/jamia/ocae129","title":"RefAI: a GPT-powered retrieval-augmented generative tool for biomedical literature recommendation and summarization","abstract":"OBJECTIVES: Precise literature recommendation and summarization are crucial for biomedical professionals. While the latest iteration of generative pretrained transformer (GPT) incorporates 2 distinct modes-real-time search and pretrained model utilization-it encounters challenges in dealing with these tasks. Specifically, the real-time search can pinpoint some relevant articles but occasionally provides fabricated papers, whereas the pretrained model excels in generating well-structured summaries but struggles to cite specific sources. In response, this study introduces RefAI, an innovative retrieval-augmented generative tool designed to synergize the strengths of large language models (LLMs) while overcoming their limitations. MATERIALS AND METHODS: RefAI utilized PubMed for systematic literature retrieval, employed a novel multivariable algorithm for article recommendation, and leveraged GPT-4 turbo for summarization. Ten queries under 2 prevalent topics (\"cancer immunotherapy and target therapy\" and \"LLMs in medicine\") were chosen as use cases and 3 established counterparts (ChatGPT-4, ScholarAI, and Gemini) as our baselines. The evaluation was conducted by 10 domain experts through standard statistical analyses for performance comparison. RESULTS: The overall performance of RefAI surpassed that of the baselines across 5 evaluated dimensions-relevance and quality for literature recommendation, accuracy, comprehensiveness, and reference integration for summarization, with the majority exhibiting statistically significant improvements (P-values <.05). DISCUSSION: RefAI demonstrated substantial improvements in literature recommendation and summarization over existing tools, addressing issues like fabricated papers, metadata inaccuracies, restricted recommendations, and poor reference integration. CONCLUSION: By augmenting LLM with external resources and a novel ranking algorithm, RefAI is uniquely capable of recommending high-quality literature and generating well-structured summaries, holding the potential to meet the critical needs of biomedical professionals in navigating and synthesizing vast amounts of scientific literature.","journal":"Journal of the American Medical Informatics Association","year":2024,"id":417845,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":60,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9578,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2024-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1205093,"name":"Jeff Zhao","orcid":null,"position":1,"is_corresponding":false},{"id":1204427,"name":"Manqi Li","orcid":"0000-0002-5206-0369","position":2,"is_corresponding":false},{"id":1048847,"name":"Yifang Dang","orcid":"0000-0002-4014-2957","position":3,"is_corresponding":false},{"id":255745,"name":"Evan Y. Yu","orcid":"0000-0002-1510-8044","position":4,"is_corresponding":false},{"id":576098,"name":"Jianfu Li","orcid":"0000-0002-9949-7007","position":5,"is_corresponding":false},{"id":1204428,"name":"Zenan Sun","orcid":"0000-0003-1577-874X","position":6,"is_corresponding":false},{"id":1205094,"name":"Usama Hussein","orcid":null,"position":7,"is_corresponding":false},{"id":1063797,"name":"Jianguo Wen","orcid":"0000-0002-3755-0044","position":8,"is_corresponding":false},{"id":1204429,"name":"Ahmed Abdelhameed","orcid":"0000-0002-9970-2517","position":9,"is_corresponding":false},{"id":254274,"name":"Junhua Mai","orcid":"0000-0001-8374-4352","position":10,"is_corresponding":false},{"id":963495,"name":"Shenduo Li","orcid":"0009-0000-1295-1960","position":11,"is_corresponding":false},{"id":492395,"name":"Yue Yu","orcid":"0000-0002-7620-6078","position":12,"is_corresponding":false},{"id":1048848,"name":"Xinyue Hu","orcid":"0009-0006-2685-951X","position":13,"is_corresponding":false},{"id":1205095,"name":"Daowei Yang","orcid":null,"position":14,"is_corresponding":false},{"id":875908,"name":"Jingna Feng","orcid":"0000-0002-0434-2513","position":15,"is_corresponding":false},{"id":367322,"name":"Zehan Li","orcid":"0000-0002-2955-6809","position":16,"is_corresponding":false},{"id":1169984,"name":"Jianping He","orcid":"0000-0002-4059-9455","position":17,"is_corresponding":false},{"id":240277,"name":"Wei Tao","orcid":"0000-0002-4277-3728","position":18,"is_corresponding":false},{"id":1204430,"name":"Tiehang Duan","orcid":"0000-0003-4323-642X","position":19,"is_corresponding":false},{"id":620393,"name":"Yanyan Lou","orcid":"0000-0001-6207-9461","position":20,"is_corresponding":false},{"id":347168,"name":"Fang Li","orcid":"0000-0001-8865-7717","position":21,"is_corresponding":false},{"id":23317,"name":"Cui Tao","orcid":"0000-0002-4267-1924","position":22,"is_corresponding":false},{"id":1169983,"name":"Yiming Li","orcid":"0009-0009-8784-1745","position":0,"is_corresponding":true}],"reference_count":26,"raw_metadata":null,"created_at":"2026-07-19T01:56:56.807779Z","pmid":"38857454","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}