{"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/training-classifiers-with-natural-language","title":"Training Classifiers with Natural Language Explanations","arxiv_id":"1805.03818","date":"2018-05-10","proceeding":"ACL 2018 7","authors":["Braden Hancock","Paroma Varma","Stephanie Wang","Martin Bringmann","Percy Liang","Christopher Ré"],"abstract":"Training accurate classifiers requires many labels, but each label provides\nonly limited information (one bit for binary classification). In this work, we\npropose BabbleLabble, a framework for training classifiers in which an\nannotator provides a natural language explanation for each labeling decision. A\nsemantic parser converts these explanations into programmatic labeling\nfunctions that generate noisy labels for an arbitrary amount of unlabeled data,\nwhich is used to train a classifier. On three relation extraction tasks, we\nfind that users are able to train classifiers with comparable F1 scores from\n5-100$\\times$ faster by providing explanations instead of just labels.\nFurthermore, given the inherent imperfection of labeling functions, we find\nthat a simple rule-based semantic parser suffices.","url_abs":"http://arxiv.org/abs/1805.03818v4","url_pdf":"http://arxiv.org/pdf/1805.03818v4.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":"training-classifiers-with-natural-language","repo_url":"https://worksheets.codalab.org/worksheets/0x900e7e41deaa4ec5b2fe41dc50594548","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"training-classifiers-with-natural-language","repo_url":"https://github.com/MurtyShikhar/ExpBERT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"binary-classification","task_name":"Binary Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.03818","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}