{"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/deep-semantic-role-labeling-with-self","title":"Deep Semantic Role Labeling with Self-Attention","arxiv_id":"1712.01586","date":"2017-12-05","proceeding":null,"authors":["Zhixing Tan","Mingxuan Wang","Jun Xie","Yidong Chen","Xiaodong Shi"],"abstract":"Semantic Role Labeling (SRL) is believed to be a crucial step towards natural\nlanguage understanding and has been widely studied. Recent years, end-to-end\nSRL with recurrent neural networks (RNN) has gained increasing attention.\nHowever, it remains a major challenge for RNNs to handle structural information\nand long range dependencies. In this paper, we present a simple and effective\narchitecture for SRL which aims to address these problems. Our model is based\non self-attention which can directly capture the relationships between two\ntokens regardless of their distance. Our single model achieves F$_1=83.4$ on\nthe CoNLL-2005 shared task dataset and F$_1=82.7$ on the CoNLL-2012 shared task\ndataset, which outperforms the previous state-of-the-art results by $1.8$ and\n$1.0$ F$_1$ score respectively. Besides, our model is computationally\nefficient, and the parsing speed is 50K tokens per second on a single Titan X\nGPU.","url_abs":"http://arxiv.org/abs/1712.01586v1","url_pdf":"http://arxiv.org/pdf/1712.01586v1.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":"deep-semantic-role-labeling-with-self","repo_url":"https://github.com/XMUNLP/Tagger","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"semantic-role-labeling","task_name":"Semantic Role Labeling"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-role-labeling-on-ontonotes","task":"Semantic Role Labeling","dataset":"OntoNotes","model":"Tan et al.","rank_in_archive_order":15,"of":17,"metrics":{"F1":"82.7"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.01586","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}