{"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/scalable-multi-domain-dialogue-state-tracking","title":"Scalable Multi-Domain Dialogue State Tracking","arxiv_id":"1712.10224","date":"2017-12-29","proceeding":null,"authors":["Abhinav Rastogi","Dilek Hakkani-Tur","Larry Heck"],"abstract":"Dialogue state tracking (DST) is a key component of task-oriented dialogue\nsystems. DST estimates the user's goal at each user turn given the interaction\nuntil then. State of the art approaches for state tracking rely on deep\nlearning methods, and represent dialogue state as a distribution over all\npossible slot values for each slot present in the ontology. Such a\nrepresentation is not scalable when the set of possible values are unbounded\n(e.g., date, time or location) or dynamic (e.g., movies or usernames).\nFurthermore, training of such models requires labeled data, where each user\nturn is annotated with the dialogue state, which makes building models for new\ndomains challenging. In this paper, we present a scalable multi-domain deep\nlearning based approach for DST. We introduce a novel framework for state\ntracking which is independent of the slot value set, and represent the dialogue\nstate as a distribution over a set of values of interest (candidate set)\nderived from the dialogue history or knowledge. Restricting these candidate\nsets to be bounded in size addresses the problem of slot-scalability.\nFurthermore, by leveraging the slot-independent architecture and transfer\nlearning, we show that our proposed approach facilitates quick adaptation to\nnew domains.","url_abs":"http://arxiv.org/abs/1712.10224v2","url_pdf":"http://arxiv.org/pdf/1712.10224v2.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":"scalable-multi-domain-dialogue-state-tracking","repo_url":"https://github.com/google-research-datasets/simulated-dialogue","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"dialogue-state-tracking","task_name":"Dialogue State Tracking"},{"task_slug":"multi-domain-dialogue-state-tracking","task_name":"Multi-domain Dialogue State Tracking"},{"task_slug":"task-oriented-dialogue-systems","task_name":"Task-Oriented Dialogue Systems"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"dst","method_name":"DST"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.10224","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}