{"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-deep-relevance-matching-model-for-ad-hoc","title":"A Deep Relevance Matching Model for Ad-hoc Retrieval","arxiv_id":"1711.08611","date":"2017-11-23","proceeding":null,"authors":["Jiafeng Guo","Yixing Fan","Qingyao Ai","W. Bruce Croft"],"abstract":"In recent years, deep neural networks have led to exciting breakthroughs in\nspeech recognition, computer vision, and natural language processing (NLP)\ntasks. However, there have been few positive results of deep models on ad-hoc\nretrieval tasks. This is partially due to the fact that many important\ncharacteristics of the ad-hoc retrieval task have not been well addressed in\ndeep models yet. Typically, the ad-hoc retrieval task is formalized as a\nmatching problem between two pieces of text in existing work using deep models,\nand treated equivalent to many NLP tasks such as paraphrase identification,\nquestion answering and automatic conversation. However, we argue that the\nad-hoc retrieval task is mainly about relevance matching while most NLP\nmatching tasks concern semantic matching, and there are some fundamental\ndifferences between these two matching tasks. Successful relevance matching\nrequires proper handling of the exact matching signals, query term importance,\nand diverse matching requirements. In this paper, we propose a novel deep\nrelevance matching model (DRMM) for ad-hoc retrieval. Specifically, our model\nemploys a joint deep architecture at the query term level for relevance\nmatching. By using matching histogram mapping, a feed forward matching network,\nand a term gating network, we can effectively deal with the three relevance\nmatching factors mentioned above. Experimental results on two representative\nbenchmark collections show that our model can significantly outperform some\nwell-known retrieval models as well as state-of-the-art deep matching models.","url_abs":"http://arxiv.org/abs/1711.08611v1","url_pdf":"http://arxiv.org/pdf/1711.08611v1.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-deep-relevance-matching-model-for-ad-hoc","repo_url":"https://github.com/EmanueleC/DRMM_repro","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"a-deep-relevance-matching-model-for-ad-hoc","repo_url":"https://github.com/faneshion/DRMM","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"a-deep-relevance-matching-model-for-ad-hoc","repo_url":"https://github.com/sebastian-hofstaetter/neural-ranking-drmm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"ad-hoc-information-retrieval","task_name":"Ad-Hoc Information Retrieval"},{"task_slug":"paraphrase-identification","task_name":"Paraphrase Identification"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/ad-hoc-information-retrieval-on-trec-robust04","task":"Ad-Hoc Information Retrieval","dataset":"TREC Robust04","model":"DRMM","rank_in_archive_order":15,"of":21,"metrics":{"MAP":"0.279","P@20":"0.382","nDCG@20":"0.431"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.08611","atlas_url":"https://app.syntology.ai/?focus=1711.08611","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}