{"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/the-web-as-a-knowledge-base-for-answering","title":"The Web as a Knowledge-base for Answering Complex Questions","arxiv_id":"1803.06643","date":"2018-03-18","proceeding":"NAACL 2018 6","authors":["Alon Talmor","Jonathan Berant"],"abstract":"Answering complex questions is a time-consuming activity for humans that\nrequires reasoning and integration of information. Recent work on reading\ncomprehension made headway in answering simple questions, but tackling complex\nquestions is still an ongoing research challenge. Conversely, semantic parsers\nhave been successful at handling compositionality, but only when the\ninformation resides in a target knowledge-base. In this paper, we present a\nnovel framework for answering broad and complex questions, assuming answering\nsimple questions is possible using a search engine and a reading comprehension\nmodel. We propose to decompose complex questions into a sequence of simple\nquestions, and compute the final answer from the sequence of answers. To\nillustrate the viability of our approach, we create a new dataset of complex\nquestions, ComplexWebQuestions, and present a model that decomposes questions\nand interacts with the web to compute an answer. We empirically demonstrate\nthat question decomposition improves performance from 20.8 precision@1 to 27.5\nprecision@1 on this new dataset.","url_abs":"http://arxiv.org/abs/1803.06643v1","url_pdf":"http://arxiv.org/pdf/1803.06643v1.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":"the-web-as-a-knowledge-base-for-answering","repo_url":"https://github.com/alontalmor/WebAsKB","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"the-web-as-a-knowledge-base-for-answering","repo_url":"https://github.com/happen2me/complex-web-questions-dataset","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"}],"methods":[],"datasets_introduced":[{"slug":"complexwebquestions","name":"ComplexWebQuestions","full_name":"ComplexWebQuestions"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.06643","atlas_url":"https://app.syntology.ai/?focus=1803.06643","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}