{"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/adversarial-training-for-multi-context-joint","title":"Adversarial training for multi-context joint entity and relation extraction","arxiv_id":"1808.06876","date":"2018-08-21","proceeding":"EMNLP 2018 10","authors":["Giannis Bekoulis","Johannes Deleu","Thomas Demeester","Chris Develder"],"abstract":"Adversarial training (AT) is a regularization method that can be used to\nimprove the robustness of neural network methods by adding small perturbations\nin the training data. We show how to use AT for the tasks of entity recognition\nand relation extraction. In particular, we demonstrate that applying AT to a\ngeneral purpose baseline model for jointly extracting entities and relations,\nallows improving the state-of-the-art effectiveness on several datasets in\ndifferent contexts (i.e., news, biomedical, and real estate data) and for\ndifferent languages (English and Dutch).","url_abs":"http://arxiv.org/abs/1808.06876v3","url_pdf":"http://arxiv.org/pdf/1808.06876v3.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":"adversarial-training-for-multi-context-joint","repo_url":"https://github.com/bekou/multihead_joint_entity_relation_extraction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"joint-entity-and-relation-extraction","task_name":"Joint Entity and Relation Extraction"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/relation-extraction-on-ace-2004","task":"Relation Extraction","dataset":"ACE 2004","model":"multi-head + AT","rank_in_archive_order":7,"of":11,"metrics":{"Cross Sentence":"No","NER Micro F1":"81.64","RE+ Micro F1":"47.45"},"uses_additional_data":false},{"leaderboard":"/sota/relation-extraction-on-ade-corpus","task":"Relation Extraction","dataset":"Adverse Drug Events (ADE) Corpus","model":"multi-head + AT","rank_in_archive_order":14,"of":15,"metrics":{"NER Macro F1":"86.73","RE+ Macro F1":"75.52"},"uses_additional_data":false},{"leaderboard":"/sota/relation-extraction-on-conll04","task":"Relation Extraction","dataset":"CoNLL04","model":"multi-head + AT","rank_in_archive_order":8,"of":16,"metrics":{"NER Macro F1":"83.6","RE+ Macro F1 ":"61.95"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.06876","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}