{"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/question-asking-as-program-generation","title":"Question Asking as Program Generation","arxiv_id":"1711.06351","date":"2017-11-16","proceeding":"NeurIPS 2017 12","authors":["Anselm Rothe","Brenden M. Lake","Todd M. Gureckis"],"abstract":"A hallmark of human intelligence is the ability to ask rich, creative, and\nrevealing questions. Here we introduce a cognitive model capable of\nconstructing human-like questions. Our approach treats questions as formal\nprograms that, when executed on the state of the world, output an answer. The\nmodel specifies a probability distribution over a complex, compositional space\nof programs, favoring concise programs that help the agent learn in the current\ncontext. We evaluate our approach by modeling the types of open-ended questions\ngenerated by humans who were attempting to learn about an ambiguous situation\nin a game. We find that our model predicts what questions people will ask, and\ncan creatively produce novel questions that were not present in the training\nset. In addition, we compare a number of model variants, finding that both\nquestion informativeness and complexity are important for producing human-like\nquestions.","url_abs":"http://arxiv.org/abs/1711.06351v1","url_pdf":"http://arxiv.org/pdf/1711.06351v1.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":"question-asking-as-program-generation","repo_url":"https://github.com/anselmrothe/eig","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"informativeness","task_name":"Informativeness"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.06351","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}