{"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/why-we-need-new-evaluation-metrics-for-nlg","title":"Why We Need New Evaluation Metrics for NLG","arxiv_id":"1707.06875","date":"2017-07-21","proceeding":"EMNLP 2017 9","authors":["Jekaterina Novikova","Ondřej Dušek","Amanda Cercas Curry","Verena Rieser"],"abstract":"The majority of NLG evaluation relies on automatic metrics, such as BLEU . In\nthis paper, we motivate the need for novel, system- and data-independent\nautomatic evaluation methods: We investigate a wide range of metrics, including\nstate-of-the-art word-based and novel grammar-based ones, and demonstrate that\nthey only weakly reflect human judgements of system outputs as generated by\ndata-driven, end-to-end NLG. We also show that metric performance is data- and\nsystem-specific. Nevertheless, our results also suggest that automatic metrics\nperform reliably at system-level and can support system development by finding\ncases where a system performs poorly.","url_abs":"http://arxiv.org/abs/1707.06875v1","url_pdf":"http://arxiv.org/pdf/1707.06875v1.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":"why-we-need-new-evaluation-metrics-for-nlg","repo_url":"https://github.com/jeknov/EMNLP_17_submission","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"nlg-evaluation","task_name":"nlg evaluation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1707.06875","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}