Abstract
Recent approaches in Incomplete Utterance Rewriting (IUR) fail to capture the source of important words, which is crucial to edit the incomplete utterance, and introduce words from irrelevant utterances. We propose a novel and effective multi-task information interaction framework including context selection, edit matrix construction, and relevance merging to capture the multi-granularity of semantic information. Benefiting from fetching the relevant utterance and figuring out the important words, our approach outperforms existing state-of-the-art models on two benchmark datasets Restoration-200K and CANAND in this field.- Anthology ID:
- 2023.findings-emnlp.168
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2576–2581
- Language:
- URL:
- https://aclanthology.org/2023.findings-emnlp.168
- DOI:
- 10.18653/v1/2023.findings-emnlp.168
- Cite (ACL):
- Haowei Du, Dinghao Zhang, Chen Li, Yang Li, and Dongyan Zhao. 2023. Multi-Granularity Information Interaction Framework for Incomplete Utterance Rewriting. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 2576–2581, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Multi-Granularity Information Interaction Framework for Incomplete Utterance Rewriting (Du et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-3/2023.findings-emnlp.168.pdf