{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/lessons-from-natural-language-inference-in","title":"Lessons from Natural Language Inference in the Clinical Domain","arxiv_id":"1808.06752","date":"2018-08-21","proceeding":"EMNLP 2018 10","authors":["Alexey Romanov","Chaitanya Shivade"],"abstract":"State of the art models using deep neural networks have become very good in\nlearning an accurate mapping from inputs to outputs. However, they still lack\ngeneralization capabilities in conditions that differ from the ones encountered\nduring training. This is even more challenging in specialized, and knowledge\nintensive domains, where training data is limited. To address this gap, we\nintroduce MedNLI - a dataset annotated by doctors, performing a natural\nlanguage inference task (NLI), grounded in the medical history of patients. We\npresent strategies to: 1) leverage transfer learning using datasets from the\nopen domain, (e.g. SNLI) and 2) incorporate domain knowledge from external data\nand lexical sources (e.g. medical terminologies). Our results demonstrate\nperformance gains using both strategies.","url_abs":"http://arxiv.org/abs/1808.06752v2","url_pdf":"http://arxiv.org/pdf/1808.06752v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"lessons-from-natural-language-inference-in","repo_url":"https://github.com/crherlihy/clinical_nli_artifacts","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"lessons-from-natural-language-inference-in","repo_url":"https://github.com/jgc128/mednli","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"lessons-from-natural-language-inference-in","repo_url":"https://github.com/jgc128/mednli_baseline","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.06752","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}