{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T13:12:40Z","timestamp":1784121160629,"version":"3.55.0"},"reference-count":32,"publisher":"Oxford University Press (OUP)","issue":"7","license":[{"start":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:00:00Z","timestamp":1750291200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Central Universities","award":["0854-53"],"award-info":[{"award-number":["0854-53"]}]},{"name":"Natural Science Foundation of Liaoning Province, China","award":["2022-BS-104"],"award-info":[{"award-number":["2022-BS-104"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>With the advancement of large language models (LLMs), the field of biomedical document-level relation extraction (BioDocRE) has encountered new opportunities. However, LLMs often face challenges such as hallucinated generation, insufficient reasoning capabilities, and a lack of interpretability when performing relation extraction tasks.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>To address these issues, we propose the SyRACT (Synergistic Retrieval Augmented Generation and Chain of Thought) framework for high precision relation extraction in biomedical documents. This framework is built around three core strategies: (i) reframing the relation extraction task as a question answering problem to better align with the processing logic of LLMs; (ii) leveraging an external database constructed from PubMed to provide LLMs with rich and reliable contextual information, thus mitigating hallucination generation; and (iii) construct a specific Chain of Thought for BioDocRE tasks, thereby enhancing the model\u2019s reasoning ability and the interpretability of its output. We validated this approach on three biomedical relation extraction datasets: CDR, GDA, and ADE. Experimental results show that the SyRACT model improves F1 scores by 11.04%, 9.10%, and 41.00% on three datasets, respectively, compared to the DocRE method, which uses standard prompts for LLMs.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>Our source code and data are available at https:\/\/github.com\/donggggxin\/SyRACT.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaf356","type":"journal-article","created":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T17:02:41Z","timestamp":1751043761000},"source":"Crossref","is-referenced-by-count":6,"title":["SyRACT: zero-shot biomedical document-level relation extraction with synergistic RAG and CoT"],"prefix":"10.1093","volume":"41","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-0267-0018","authenticated-orcid":false,"given":"Xin","family":"Dong","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Dalian Minzu University , Liaoning 116600,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0876-5126","authenticated-orcid":false,"given":"Di","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Dalian Minzu University , Liaoning 116600,","place":["China"]},{"name":"School of Computer Science and Technology, Dalian University of Technology , Liaoning 116024,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiana","family":"Meng","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Dalian Minzu University , Liaoning 116600,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bocheng","family":"Guo","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Dalian Minzu University , Liaoning 116600,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongfei","family":"Lin","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Dalian University of Technology , Liaoning 116024,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,6,19]]},"reference":[{"key":"2025070907500783300_btaf356-B1","first-page":"318"},{"key":"2025070907500783300_btaf356-B2","doi-asserted-by":"crossref","first-page":"109184","DOI":"10.1016\/j.knosys.2022.109184","article-title":"On the form of parsed sentences for relation extraction","volume":"251","author":"Chen","year":"2022","journal-title":"Knowl-Based Syst"},{"key":"2025070907500783300_btaf356-B3","first-page":"6491"},{"key":"2025070907500783300_btaf356-B4","doi-asserted-by":"crossref","first-page":"127881","DOI":"10.1016\/j.neucom.2024.127881","article-title":"Few-shot biomedical relation extraction using data augmentation and domain information","volume":"595","author":"Guo","year":"2024","journal-title":"Neurocomputing"},{"key":"2025070907500783300_btaf356-B5","doi-asserted-by":"crossref","first-page":"885","DOI":"10.1016\/j.jbi.2012.04.008","article-title":"Development of a benchmark corpus to support the automatic extraction of drug-related adverse effects from medical case reports","volume":"45","author":"Gurulingappa","year":"2012","journal-title":"J Biomed Inform"},{"key":"2025070907500783300_btaf356-B6","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1016\/j.aiopen.2022.11.003","article-title":"PTR: prompt tuning with rules for text classification","volume":"3","author":"Han","year":"2022","journal-title":"AI Open"},{"key":"2025070907500783300_btaf356-B7","first-page":"2418"},{"key":"2025070907500783300_btaf356-B8","first-page":"2820"},{"key":"2025070907500783300_btaf356-B9","first-page":"22199","article-title":"Large language models are zero-shot reasoners","volume":"35","author":"Kojima","year":"2022","journal-title":"Adv Neural Inform Process Syst"},{"key":"2025070907500783300_btaf356-B10","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1038\/s41746-024-01091-y","article-title":"Optimization of hepatological clinical guidelines interpretation by large language models: a retrieval augmented generation-based framework","volume":"7","author":"Kresevic","year":"2024","journal-title":"NPJ Digit Med"},{"key":"2025070907500783300_btaf356-B11","first-page":"2835"},{"key":"2025070907500783300_btaf356-B12","first-page":"9459","article-title":"Retrieval-augmented generation for knowledge-intensive NLP tasks","volume":"33","author":"Lewis","year":"2020","journal-title":"Adv Neural Inform Process Syst"},{"key":"2025070907500783300_btaf356-B13","first-page":"6877"},{"key":"2025070907500783300_btaf356-B14","doi-asserted-by":"crossref","first-page":"baw068","DOI":"10.1093\/database\/baw068","article-title":"Biocreative V CDR task corpus: a resource for chemical disease relation extraction","volume":"2016","author":"Li","year":"2016","journal-title":"Database"},{"key":"2025070907500783300_btaf356-B15","first-page":"5495"},{"key":"2025070907500783300_btaf356-B16","first-page":"2334"},{"key":"2025070907500783300_btaf356-B17","first-page":"1971"},{"key":"2025070907500783300_btaf356-B18","first-page":"11375"},{"key":"2025070907500783300_btaf356-B19","first-page":"4407"},{"key":"2025070907500783300_btaf356-B20"},{"key":"2025070907500783300_btaf356-B21","doi-asserted-by":"crossref","first-page":"59","DOI":"10.7326\/M19-2548","article-title":"Should health care demand interpretable artificial intelligence or accept \u201cblack box\u201d medicine?","volume":"172","author":"Wang","year":"2020","journal-title":"Ann Intern Med"},{"key":"2025070907500783300_btaf356-B22","first-page":"24824","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv Neural Inform Process Syst"},{"key":"2025070907500783300_btaf356-B23","first-page":"272"},{"key":"2025070907500783300_btaf356-B24","first-page":"2395"},{"key":"2025070907500783300_btaf356-B25","first-page":"2904"},{"key":"2025070907500783300_btaf356-B26","doi-asserted-by":"crossref","first-page":"101574","DOI":"10.1016\/j.csl.2023.101574","article-title":"Document-level relation extraction with entity mentions deep attention","volume":"84","author":"Xu","year":"2024","journal-title":"Comput Speech Lang"},{"key":"2025070907500783300_btaf356-B27","first-page":"211"},{"key":"2025070907500783300_btaf356-B28","first-page":"16714"},{"key":"2025070907500783300_btaf356-B29"},{"key":"2025070907500783300_btaf356-B30","doi-asserted-by":"crossref","first-page":"104459","DOI":"10.1016\/j.jbi.2023.104459","article-title":"Biomedical document relation extraction with prompt learning and KNN","volume":"145","author":"Zhao","year":"2023","journal-title":"J Biomed Inform"},{"key":"2025070907500783300_btaf356-B31","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jbi.2018.05.001","article-title":"An effective neural model extracting document level chemical-induced disease relations from biomedical literature","volume":"83","author":"Zheng","year":"2018","journal-title":"J Biomed Inform"},{"key":"2025070907500783300_btaf356-B32","first-page":"5259"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btaf356\/63528032\/btaf356.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/7\/btaf356\/63528032\/btaf356.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/7\/btaf356\/63528032\/btaf356.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,9]],"date-time":"2025-07-09T11:50:18Z","timestamp":1752061818000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btaf356\/8169328"}},"subtitle":[],"editor":[{"given":"Zhiyong","family":"Lu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,6,19]]},"references-count":32,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2025,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaf356","relation":{},"ISSN":["1367-4811"],"issn-type":[{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,7]]},"published":{"date-parts":[[2025,6,19]]},"article-number":"btaf356"}}