Abstract
There has been significant interest in zero and few-shot learning for dialogue state tracking (DST) due to the high cost of collecting and annotating task-oriented dialogues. Recent work has demonstrated that in-context learning requires very little data and zero parameter updates, and even outperforms trained methods in the few-shot setting. We propose RefPyDST, which advances the state of the art with three advancements to in-context learning for DST.First, we formulate DST as a Python programming task, explicitly modeling language coreference as variable reference in Python. Second, since in-context learning depends highly on the context examples, we propose a method to retrieve a diverse set of relevant examples to improve performance. Finally, we introduce a novel re-weighting method during decoding that takes into account probabilities of competing surface forms, and produces a more accurate dialogue state prediction. We evaluate our approach using MultiWOZ and achieve state-of-the-art multi-domain joint-goal accuracy in zero and few-shot settings.- Anthology ID:
- 2023.findings-acl.344
- Original:
- 2023.findings-acl.344v1
- Version 2:
- 2023.findings-acl.344v2
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2023
- Month:
- July
- Year:
- 2023
- Address:
- Toronto, Canada
- Editors:
- Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5570–5585
- Language:
- URL:
- https://aclanthology.org/2023.findings-acl.344
- DOI:
- 10.18653/v1/2023.findings-acl.344
- Cite (ACL):
- Brendan King and Jeffrey Flanigan. 2023. Diverse Retrieval-Augmented In-Context Learning for Dialogue State Tracking. In Findings of the Association for Computational Linguistics: ACL 2023, pages 5570–5585, Toronto, Canada. Association for Computational Linguistics.
- Cite (Informal):
- Diverse Retrieval-Augmented In-Context Learning for Dialogue State Tracking (King & Flanigan, Findings 2023)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2023.findings-acl.344.pdf