{"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/program-induction-by-rationale-generation","title":"Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems","arxiv_id":"1705.04146","date":"2017-05-11","proceeding":null,"authors":["Wang Ling","Dani Yogatama","Chris Dyer","Phil Blunsom"],"abstract":"Solving algebraic word problems requires executing a series of arithmetic\noperations---a program---to obtain a final answer. However, since programs can\nbe arbitrarily complicated, inducing them directly from question-answer pairs\nis a formidable challenge. To make this task more feasible, we solve these\nproblems by generating answer rationales, sequences of natural language and\nhuman-readable mathematical expressions that derive the final answer through a\nseries of small steps. Although rationales do not explicitly specify programs,\nthey provide a scaffolding for their structure via intermediate milestones. To\nevaluate our approach, we have created a new 100,000-sample dataset of\nquestions, answers and rationales. Experimental results show that indirect\nsupervision of program learning via answer rationales is a promising strategy\nfor inducing arithmetic programs.","url_abs":"http://arxiv.org/abs/1705.04146v3","url_pdf":"http://arxiv.org/pdf/1705.04146v3.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":"program-induction-by-rationale-generation","repo_url":"https://github.com/deepmind/AQuA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"program-induction","task_name":"Program induction"}],"methods":[],"datasets_introduced":[{"slug":"aqua-rat","name":"AQUA-RAT","full_name":"Algebra Question Answering with Rationales"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.04146","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}