Abstract
In a multi-turn knowledge-grounded dialog, the difference between the knowledge selected at different turns usually provides potential clues to knowledge selection, which has been largely neglected in previous research. In this paper, we propose a difference-aware knowledge selection method. It first computes the difference between the candidate knowledge sentences provided at the current turn and those chosen in the previous turns. Then, the differential information is fused with or disentangled from the contextual information to facilitate final knowledge selection. Automatic, human observational, and interactive evaluation shows that our method is able to select knowledge more accurately and generate more informative responses, significantly outperforming the state-of-the-art baselines.- Anthology ID:
- 2020.findings-emnlp.11
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2020
- Month:
- November
- Year:
- 2020
- Address:
- Online
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 115–125
- Language:
- URL:
- https://aclanthology.org/2020.findings-emnlp.11
- DOI:
- 10.18653/v1/2020.findings-emnlp.11
- Cite (ACL):
- Chujie Zheng, Yunbo Cao, Daxin Jiang, and Minlie Huang. 2020. Difference-aware Knowledge Selection for Knowledge-grounded Conversation Generation. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 115–125, Online. Association for Computational Linguistics.
- Cite (Informal):
- Difference-aware Knowledge Selection for Knowledge-grounded Conversation Generation (Zheng et al., Findings 2020)
- PDF:
- https://preview.aclanthology.org/nodalida-main-page/2020.findings-emnlp.11.pdf
- Code
- chujiezheng/DiffKS
- Data
- Holl-E, Wizard of Wikipedia