{"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/empower-sequence-labeling-with-task-aware","title":"Empower Sequence Labeling with Task-Aware Neural Language Model","arxiv_id":"1709.04109","date":"2017-09-13","proceeding":null,"authors":["Liyuan Liu","Jingbo Shang","Frank F. Xu","Xiang Ren","Huan Gui","Jian Peng","Jiawei Han"],"abstract":"Linguistic sequence labeling is a general modeling approach that encompasses\na variety of problems, such as part-of-speech tagging and named entity\nrecognition. Recent advances in neural networks (NNs) make it possible to build\nreliable models without handcrafted features. However, in many cases, it is\nhard to obtain sufficient annotations to train these models. In this study, we\ndevelop a novel neural framework to extract abundant knowledge hidden in raw\ntexts to empower the sequence labeling task. Besides word-level knowledge\ncontained in pre-trained word embeddings, character-aware neural language\nmodels are incorporated to extract character-level knowledge. Transfer learning\ntechniques are further adopted to mediate different components and guide the\nlanguage model towards the key knowledge. Comparing to previous methods, these\ntask-specific knowledge allows us to adopt a more concise model and conduct\nmore efficient training. Different from most transfer learning methods, the\nproposed framework does not rely on any additional supervision. It extracts\nknowledge from self-contained order information of training sequences.\nExtensive experiments on benchmark datasets demonstrate the effectiveness of\nleveraging character-level knowledge and the efficiency of co-training. For\nexample, on the CoNLL03 NER task, model training completes in about 6 hours on\na single GPU, reaching F1 score of 91.71$\\pm$0.10 without using any extra\nannotation.","url_abs":"http://arxiv.org/abs/1709.04109v4","url_pdf":"http://arxiv.org/pdf/1709.04109v4.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":"empower-sequence-labeling-with-task-aware","repo_url":"https://github.com/LiyuanLucasLiu/LM-LSTM-CRF","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"empower-sequence-labeling-with-task-aware","repo_url":"https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Sequence-Labeling","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"empower-sequence-labeling-with-task-aware","repo_url":"https://github.com/zysite/post","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"cg","task_name":"NER"},{"task_slug":"named-entity-recognition-1","task_name":"Named Entity Recognition"},{"task_slug":"named-entity-recognition-ner","task_name":"Named Entity Recognition (NER)"},{"task_slug":"part-of-speech-tagging","task_name":"Part-Of-Speech Tagging"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"},{"task_slug":"named-entity-recognition","task_name":"named-entity-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/named-entity-recognition-ner-on-conll-2003","task":"Named Entity Recognition (NER)","dataset":"CoNLL 2003 (English)","model":"LM-LSTM-CRF","rank_in_archive_order":66,"of":73,"metrics":{"F1":"91.24"},"uses_additional_data":false},{"leaderboard":"/sota/part-of-speech-tagging-on-penn-treebank","task":"Part-Of-Speech Tagging","dataset":"Penn Treebank","model":"LM-LSTM-CRF","rank_in_archive_order":13,"of":20,"metrics":{"Accuracy":"97.53"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.04109","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}