{"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/a-question-entailment-approach-to-question","title":"A Question-Entailment Approach to Question Answering","arxiv_id":"1901.08079","date":"2019-01-23","proceeding":null,"authors":["Asma Ben Abacha","Dina Demner-Fushman"],"abstract":"One of the challenges in large-scale information retrieval (IR) is to develop\nfine-grained and domain-specific methods to answer natural language questions.\nDespite the availability of numerous sources and datasets for answer retrieval,\nQuestion Answering (QA) remains a challenging problem due to the difficulty of\nthe question understanding and answer extraction tasks. One of the promising\ntracks investigated in QA is to map new questions to formerly answered\nquestions that are `similar'. In this paper, we propose a novel QA approach\nbased on Recognizing Question Entailment (RQE) and we describe the QA system\nand resources that we built and evaluated on real medical questions. First, we\ncompare machine learning and deep learning methods for RQE using different\nkinds of datasets, including textual inference, question similarity and\nentailment in both the open and clinical domains. Second, we combine IR models\nwith the best RQE method to select entailed questions and rank the retrieved\nanswers. To study the end-to-end QA approach, we built the MedQuAD collection\nof 47,457 question-answer pairs from trusted medical sources, that we introduce\nand share in the scope of this paper. Following the evaluation process used in\nTREC 2017 LiveQA, we find that our approach exceeds the best results of the\nmedical task with a 29.8% increase over the best official score. The evaluation\nresults also support the relevance of question entailment for QA and highlight\nthe effectiveness of combining IR and RQE for future QA efforts. Our findings\nalso show that relying on a restricted set of reliable answer sources can bring\na substantial improvement in medical QA.","url_abs":"http://arxiv.org/abs/1901.08079v1","url_pdf":"http://arxiv.org/pdf/1901.08079v1.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":"a-question-entailment-approach-to-question","repo_url":"https://github.com/abachaa/MedQuAD","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"a-question-entailment-approach-to-question","repo_url":"https://github.com/abachaa/LiveQA_MedicalTask_TREC2017","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"a-question-entailment-approach-to-question","repo_url":"https://github.com/stormieGal/mediLearn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"question-similarity","task_name":"Question Similarity"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[{"slug":"medquad","name":"MedQuAD","full_name":"Medical Question Answering Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1901.08079","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}