{"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/evaluation-discrepancy-discovery-a-sentence","title":"Evaluation Discrepancy Discovery: A Sentence Compression Case-study","arxiv_id":"2101.09079","date":"2021-01-22","proceeding":null,"authors":["Yevgeniy Puzikov"],"abstract":"Reliable evaluation protocols are of utmost importance for reproducible NLP research. In this work, we show that sometimes neither metric nor conventional human evaluation is sufficient to draw conclusions about system performance. Using sentence compression as an example task, we demonstrate how a system can game a well-established dataset to achieve state-of-the-art results. In contrast with the results reported in previous work that showed correlation between human judgements and metric scores, our manual analysis of state-of-the-art system outputs demonstrates that high metric scores may only indicate a better fit to the data, but not better outputs, as perceived by humans.","url_abs":"https://arxiv.org/abs/2101.09079v1","url_pdf":"https://arxiv.org/pdf/2101.09079v1.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":"evaluation-discrepancy-discovery-a-sentence","repo_url":"https://github.com/UKPLab/arxiv2021-evaluation-discrepancy-nsc","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"sentence-compression","task_name":"Sentence Compression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}