{"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/value-based-search-in-execution-space-for","title":"Value-based Search in Execution Space for Mapping Instructions to Programs","arxiv_id":"1811.01090","date":"2018-11-02","proceeding":"NAACL 2019 6","authors":["Dor Muhlgay","Jonathan Herzig","Jonathan Berant"],"abstract":"Training models to map natural language instructions to programs given target\nworld supervision only requires searching for good programs at training time.\nSearch is commonly done using beam search in the space of partial programs or\nprogram trees, but as the length of the instructions grows finding a good\nprogram becomes difficult. In this work, we propose a search algorithm that\nuses the target world state, known at training time, to train a critic network\nthat predicts the expected reward of every search state. We then score search\nstates on the beam by interpolating their expected reward with the likelihood\nof programs represented by the search state. Moreover, we search not in the\nspace of programs but in a more compressed state of program executions,\naugmented with recent entities and actions. On the SCONE dataset, we show that\nour algorithm dramatically improves performance on all three domains compared\nto standard beam search and other baselines.","url_abs":"http://arxiv.org/abs/1811.01090v2","url_pdf":"http://arxiv.org/pdf/1811.01090v2.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":"value-based-search-in-execution-space-for","repo_url":"https://gitlab.com/tau-nlp/vbsix-lang2program","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.01090","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}