{"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/towards-exploiting-background-knowledge-for","title":"Towards Exploiting Background Knowledge for Building Conversation Systems","arxiv_id":"1809.08205","date":"2018-09-21","proceeding":"EMNLP 2018 10","authors":["Nikita Moghe","Siddhartha Arora","Suman Banerjee","Mitesh M. Khapra"],"abstract":"Existing dialog datasets contain a sequence of utterances and responses\nwithout any explicit background knowledge associated with them. This has\nresulted in the development of models which treat conversation as a\nsequence-to-sequence generation task i.e, given a sequence of utterances\ngenerate the response sequence). This is not only an overly simplistic view of\nconversation but it is also emphatically different from the way humans converse\nby heavily relying on their background knowledge about the topic (as opposed to\nsimply relying on the previous sequence of utterances). For example, it is\ncommon for humans to (involuntarily) produce utterances which are copied or\nsuitably modified from background articles they have read about the topic. To\nfacilitate the development of such natural conversation models which mimic the\nhuman process of conversing, we create a new dataset containing movie chats\nwherein each response is explicitly generated by copying and/or modifying\nsentences from unstructured background knowledge such as plots, comments and\nreviews about the movie. We establish baseline results on this dataset (90K\nutterances from 9K conversations) using three different models: (i) pure\ngeneration based models which ignore the background knowledge (ii) generation\nbased models which learn to copy information from the background knowledge when\nrequired and (iii) span prediction based models which predict the appropriate\nresponse span in the background knowledge.","url_abs":"http://arxiv.org/abs/1809.08205v1","url_pdf":"http://arxiv.org/pdf/1809.08205v1.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":"towards-exploiting-background-knowledge-for","repo_url":"https://github.com/nikitacs16/Holl-E","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"articles","task_name":"Articles"}],"methods":[],"datasets_introduced":[{"slug":"holl-e","name":"Holl-E","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.08205","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}