{"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/optimizing-differentiable-relaxations-of","title":"Optimizing Differentiable Relaxations of Coreference Evaluation Metrics","arxiv_id":"1704.04451","date":"2017-04-14","proceeding":"CONLL 2017 8","authors":["Phong Le","Ivan Titov"],"abstract":"Coreference evaluation metrics are hard to optimize directly as they are\nnon-differentiable functions, not easily decomposable into elementary\ndecisions. Consequently, most approaches optimize objectives only indirectly\nrelated to the end goal, resulting in suboptimal performance. Instead, we\npropose a differentiable relaxation that lends itself to gradient-based\noptimisation, thus bypassing the need for reinforcement learning or heuristic\nmodification of cross-entropy. We show that by modifying the training objective\nof a competitive neural coreference system, we obtain a substantial gain in\nperformance. This suggests that our approach can be regarded as a viable\nalternative to using reinforcement learning or more computationally expensive\nimitation learning.","url_abs":"http://arxiv.org/abs/1704.04451v3","url_pdf":"http://arxiv.org/pdf/1704.04451v3.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":"optimizing-differentiable-relaxations-of","repo_url":"https://github.com/lephong/diffmetric_coref","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"imitation-learning","task_name":"Imitation Learning"},{"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":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}