Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encoder
Bin Sun, Shaoxiong Feng, Yiwei Li, Weichao Wang, Fei Mi, Yitong Li, Kan Li
Abstract
Complex dialogue mappings (CDM), including one-to-many and many-to-one mappings, tend to make dialogue models generate incoherent or dull responses, and modeling these mappings remains a huge challenge for neural dialogue systems. To alleviate these problems, methods like introducing external information, reconstructing the optimization function, and manipulating data samples are proposed, while they primarily focus on avoiding training with CDM, inevitably weakening the model’s ability of understanding CDM in human conversations and limiting further improvements in model performance. This paper proposes a Sentence Semantic Segmentation guided Conditional Variational Auto-Encoder (SegCVAE) method which can model and take advantages of the CDM data. Specifically, to tackle the incoherent problem caused by one-to-many, SegCVAE uses response-related prominent semantics to constrained the latent variable. To mitigate the non-diverse problem brought by many-to-one, SegCVAE segments multiple prominent semantics to enrich the latent variables. Three novel components, Internal Separation, External Guidance, and Semantic Norms, are proposed to achieve SegCVAE. On dialogue generation tasks, both the automatic and human evaluation results show that SegCVAE achieves new state-of-the-art performance.- Anthology ID:
- 2022.findings-emnlp.305
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2022
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Editors:
- Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4140–4153
- Language:
- URL:
- https://aclanthology.org/2022.findings-emnlp.305
- DOI:
- 10.18653/v1/2022.findings-emnlp.305
- Cite (ACL):
- Bin Sun, Shaoxiong Feng, Yiwei Li, Weichao Wang, Fei Mi, Yitong Li, and Kan Li. 2022. Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encoder. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 4140–4153, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encoder (Sun et al., Findings 2022)
- PDF:
- https://preview.aclanthology.org/naacl-24-ws-corrections/2022.findings-emnlp.305.pdf