{"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/a-simple-fast-diverse-decoding-algorithm-for","title":"A Simple, Fast Diverse Decoding Algorithm for Neural Generation","arxiv_id":"1611.08562","date":"2016-11-25","proceeding":null,"authors":["Jiwei Li","Will Monroe","Dan Jurafsky"],"abstract":"In this paper, we propose a simple, fast decoding algorithm that fosters\ndiversity in neural generation. The algorithm modifies the standard beam search\nalgorithm by adding an inter-sibling ranking penalty, favoring choosing\nhypotheses from diverse parents. We evaluate the proposed model on the tasks of\ndialogue response generation, abstractive summarization and machine\ntranslation. We find that diverse decoding helps across all tasks, especially\nthose for which reranking is needed.\n  We further propose a variation that is capable of automatically adjusting its\ndiversity decoding rates for different inputs using reinforcement learning\n(RL). We observe a further performance boost from this RL technique. This paper\nincludes material from the unpublished script \"Mutual Information and Diverse\nDecoding Improve Neural Machine Translation\" (Li and Jurafsky, 2016).","url_abs":"http://arxiv.org/abs/1611.08562v2","url_pdf":"http://arxiv.org/pdf/1611.08562v2.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":"a-simple-fast-diverse-decoding-algorithm-for","repo_url":"https://github.com/TaiseiAso/BiGruAttEncDec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"abstractive-text-summarization","task_name":"Abstractive Text Summarization"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reranking","task_name":"Reranking"},{"task_slug":"response-generation","task_name":"Response Generation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1611.08562","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}