{"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/joint-entity-linking-with-deep-reinforcement","title":"Joint Entity Linking with Deep Reinforcement Learning","arxiv_id":"1902.00330","date":"2019-02-01","proceeding":null,"authors":["Zheng Fang","Yanan Cao","Dongjie Zhang","Qian Li","Zhen-Yu Zhang","Yanbing Liu"],"abstract":"Entity linking is the task of aligning mentions to corresponding entities in\na given knowledge base. Previous studies have highlighted the necessity for\nentity linking systems to capture the global coherence. However, there are two\ncommon weaknesses in previous global models. First, most of them calculate the\npairwise scores between all candidate entities and select the most relevant\ngroup of entities as the final result. In this process, the consistency among\nwrong entities as well as that among right ones are involved, which may\nintroduce noise data and increase the model complexity. Second, the cues of\npreviously disambiguated entities, which could contribute to the disambiguation\nof the subsequent mentions, are usually ignored by previous models. To address\nthese problems, we convert the global linking into a sequence decision problem\nand propose a reinforcement learning model which makes decisions from a global\nperspective. Our model makes full use of the previous referred entities and\nexplores the long-term influence of current selection on subsequent decisions.\nWe conduct experiments on different types of datasets, the results show that\nour model outperforms state-of-the-art systems and has better generalization\nperformance.","url_abs":"http://arxiv.org/abs/1902.00330v1","url_pdf":"http://arxiv.org/pdf/1902.00330v1.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":[],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"entity-disambiguation","task_name":"Entity Disambiguation"},{"task_slug":"entity-linking","task_name":"Entity Linking"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/entity-disambiguation-on-aida-conll","task":"Entity Disambiguation","dataset":"AIDA-CoNLL","model":"Fang et al. (2019) (et al., [2019e])","rank_in_archive_order":6,"of":20,"metrics":{"In-KB Accuracy":"94.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.00330","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}