{"doi":"10.1038/s41746-024-01239-w","title":"Closing the gap between open source and commercial large language models for medical evidence summarization","abstract":"Large language models (LLMs) hold great promise in summarizing medical evidence. Most recent studies focus on the application of proprietary LLMs. Using proprietary LLMs introduces multiple risk factors, including a lack of transparency and vendor dependency. While open-source LLMs allow better transparency and customization, their performance falls short compared to the proprietary ones. In this study, we investigated to what extent fine-tuning open-source LLMs can further improve their performance. Utilizing a benchmark dataset, MedReview, consisting of 8161 pairs of systematic reviews and summaries, we fine-tuned three broadly-used, open-sourced LLMs, namely PRIMERA, LongT5, and Llama-2. Overall, the performance of open-source models was all improved after fine-tuning. The performance of fine-tuned LongT5 is close to GPT-3.5 with zero-shot settings. Furthermore, smaller fine-tuned models sometimes even demonstrated superior performance compared to larger zero-shot models. The above trends of improvement were manifested in both a human evaluation and a larger-scale GPT4-simulated evaluation.","journal":"npj Digital Medicine","year":2024,"id":417324,"datarank":0.6476232170304466,"base_score":4.31748811353631,"endowment":4.31748811353631,"self_citation_contribution":0.6476232170304466,"citation_network_contribution":0.0,"self_endowment_contribution":0.6476232170304466,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":74,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9527,"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":1014353,"name":"Qiao Jin","orcid":"0000-0002-1268-7239","position":1,"is_corresponding":false},{"id":1194207,"name":"Yiliang Zhou","orcid":"0009-0002-7457-7075","position":2,"is_corresponding":false},{"id":1059354,"name":"Song Wang","orcid":"0000-0002-8224-0424","position":3,"is_corresponding":false},{"id":665203,"name":"Betina Idnay","orcid":"0000-0002-4318-5987","position":4,"is_corresponding":false},{"id":966902,"name":"Hanley Ong","orcid":"0009-0004-9395-6446","position":5,"is_corresponding":false},{"id":672602,"name":"Elizabeth Park","orcid":"0000-0002-5277-349X","position":6,"is_corresponding":false},{"id":356474,"name":"Jordan G. Nestor","orcid":"0000-0003-1418-3103","position":7,"is_corresponding":false},{"id":920368,"name":"Matthew E. Spotnitz","orcid":null,"position":8,"is_corresponding":false},{"id":646886,"name":"Ali Soroush","orcid":"0000-0001-6900-5596","position":9,"is_corresponding":false},{"id":278959,"name":"Thomas R. Campion","orcid":"0000-0001-7624-769X","position":10,"is_corresponding":false},{"id":45332,"name":"ZHIYONG LU","orcid":"0000-0001-9998-916X","position":11,"is_corresponding":false},{"id":2012,"name":"Chunhua Weng","orcid":"0000-0002-9624-0214","position":12,"is_corresponding":false},{"id":85506,"name":"Yifan Peng","orcid":"0000-0001-9309-8331","position":13,"is_corresponding":false},{"id":721849,"name":"Gongbo Zhang","orcid":"0009-0001-0077-3615","position":0,"is_corresponding":true}],"reference_count":37,"raw_metadata":{"citation_network_status":"fetched"},"created_at":"2026-07-19T01:56:50.743370Z","pmid":"39251804","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":[]}