Abstract
We introduce AARGH, an end-to-end task-oriented dialog system combining retrieval and generative approaches in a single model, aiming at improving dialog management and lexical diversity of outputs. The model features a new response selection method based on an action-aware training objective and a simplified single-encoder retrieval architecture which allow us to build an end-to-end retrieval-enhanced generation model where retrieval and generation share most of the parameters. On the MultiWOZ dataset, we show that our approach produces more diverse outputs while maintaining or improving state tracking and context-to-response generation performance, compared to state-of-the-art baselines.- Anthology ID:
- 2022.sigdial-1.29
- Volume:
- Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue
- Month:
- September
- Year:
- 2022
- Address:
- Edinburgh, UK
- Venue:
- SIGDIAL
- SIG:
- SIGDIAL
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 283–297
- Language:
- URL:
- https://aclanthology.org/2022.sigdial-1.29
- DOI:
- Cite (ACL):
- Tomáš Nekvinda and Ondřej Dušek. 2022. AARGH! End-to-end Retrieval-Generation for Task-Oriented Dialog. In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 283–297, Edinburgh, UK. Association for Computational Linguistics.
- Cite (Informal):
- AARGH! End-to-end Retrieval-Generation for Task-Oriented Dialog (Nekvinda & Dušek, SIGDIAL 2022)
- PDF:
- https://preview.aclanthology.org/remove-xml-comments/2022.sigdial-1.29.pdf
- Code
- tomiinek/aargh
- Data
- MultiWOZ