{"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/referenceless-quality-estimation-for-natural","title":"Referenceless Quality Estimation for Natural Language Generation","arxiv_id":"1708.01759","date":"2017-08-05","proceeding":null,"authors":["Ondřej Dušek","Jekaterina Novikova","Verena Rieser"],"abstract":"Traditional automatic evaluation measures for natural language generation\n(NLG) use costly human-authored references to estimate the quality of a system\noutput. In this paper, we propose a referenceless quality estimation (QE)\napproach based on recurrent neural networks, which predicts a quality score for\na NLG system output by comparing it to the source meaning representation only.\nOur method outperforms traditional metrics and a constant baseline in most\nrespects; we also show that synthetic data helps to increase correlation\nresults by 21% compared to the base system. Our results are comparable to\nresults obtained in similar QE tasks despite the more challenging setting.","url_abs":"http://arxiv.org/abs/1708.01759v1","url_pdf":"http://arxiv.org/pdf/1708.01759v1.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":"referenceless-quality-estimation-for-natural","repo_url":"https://github.com/tuetschek/ratpred","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.01759","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}