{"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/word-level-loss-extensions-for-neural","title":"Word-Level Loss Extensions for Neural Temporal Relation Classification","arxiv_id":"1808.02374","date":"2018-08-07","proceeding":"COLING 2018 8","authors":["Artuur Leeuwenberg","Marie-Francine Moens"],"abstract":"Unsupervised pre-trained word embeddings are used effectively for many tasks\nin natural language processing to leverage unlabeled textual data. Often these\nembeddings are either used as initializations or as fixed word representations\nfor task-specific classification models. In this work, we extend our\nclassification model's task loss with an unsupervised auxiliary loss on the\nword-embedding level of the model. This is to ensure that the learned word\nrepresentations contain both task-specific features, learned from the\nsupervised loss component, and more general features learned from the\nunsupervised loss component. We evaluate our approach on the task of temporal\nrelation extraction, in particular, narrative containment relation extraction\nfrom clinical records, and show that continued training of the embeddings on\nthe unsupervised objective together with the task objective gives better\ntask-specific embeddings, and results in an improvement over the state of the\nart on the THYME dataset, using only a general-domain part-of-speech tagger as\nlinguistic resource.","url_abs":"http://arxiv.org/abs/1808.02374v1","url_pdf":"http://arxiv.org/pdf/1808.02374v1.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":"word-level-loss-extensions-for-neural","repo_url":"https://github.com/tuur/WLLETlinkClassification","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-classification","task_name":"Relation Classification"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"},{"task_slug":"temporal-relation-classification","task_name":"Temporal Relation Classification"},{"task_slug":"temporal-relation-extraction","task_name":"Temporal Relation Extraction"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}