{"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/sequence-to-sequence-learning-for-task","title":"Sequence-to-Sequence Learning for Task-oriented Dialogue with Dialogue State Representation","arxiv_id":"1806.04441","date":"2018-06-12","proceeding":"COLING 2018 8","authors":["Haoyang Wen","Yijia Liu","Wanxiang Che","Libo Qin","Ting Liu"],"abstract":"Classic pipeline models for task-oriented dialogue system require explicit\nmodeling the dialogue states and hand-crafted action spaces to query a\ndomain-specific knowledge base. Conversely, sequence-to-sequence models learn\nto map dialogue history to the response in current turn without explicit\nknowledge base querying. In this work, we propose a novel framework that\nleverages the advantages of classic pipeline and sequence-to-sequence models.\nOur framework models a dialogue state as a fixed-size distributed\nrepresentation and use this representation to query a knowledge base via an\nattention mechanism. Experiment on Stanford Multi-turn Multi-domain\nTask-oriented Dialogue Dataset shows that our framework significantly\noutperforms other sequence-to-sequence based baseline models on both automatic\nand human evaluation.","url_abs":"http://arxiv.org/abs/1806.04441v1","url_pdf":"http://arxiv.org/pdf/1806.04441v1.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":[],"tasks":[{"task_slug":"task-oriented-dialogue-systems","task_name":"Task-Oriented Dialogue Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/task-oriented-dialogue-systems-on-kvret","task":"Task-Oriented Dialogue Systems","dataset":"KVRET","model":"DSR","rank_in_archive_order":7,"of":10,"metrics":{"BLEU":"12.7","Entity F1":"51.9"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.04441","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}