{"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/towards-better-ud-parsing-deep-contextualized","title":"Towards Better UD Parsing: Deep Contextualized Word Embeddings, Ensemble, and Treebank Concatenation","arxiv_id":"1807.03121","date":"2018-07-09","proceeding":"CONLL 2018 10","authors":["Wanxiang Che","Yijia Liu","Yuxuan Wang","Bo Zheng","Ting Liu"],"abstract":"This paper describes our system (HIT-SCIR) submitted to the CoNLL 2018 shared\ntask on Multilingual Parsing from Raw Text to Universal Dependencies. We base\nour submission on Stanford's winning system for the CoNLL 2017 shared task and\nmake two effective extensions: 1) incorporating deep contextualized word\nembeddings into both the part of speech tagger and parser; 2) ensembling\nparsers trained with different initialization. We also explore different ways\nof concatenating treebanks for further improvements. Experimental results on\nthe development data show the effectiveness of our methods. In the final\nevaluation, our system was ranked first according to LAS (75.84%) and\noutperformed the other systems by a large margin.","url_abs":"http://arxiv.org/abs/1807.03121v3","url_pdf":"http://arxiv.org/pdf/1807.03121v3.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":"towards-better-ud-parsing-deep-contextualized","repo_url":"https://github.com/HIT-SCIR/ELMoForManyLangs","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/dependency-parsing-on-universal-dependencies","task":"Dependency Parsing","dataset":"Universal Dependencies","model":"HIT-SCIR","rank_in_archive_order":3,"of":6,"metrics":{"LAS":"75.84"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.03121","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}