Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning
Yanan Wu, Zhiyuan Zeng, Keqing He, Yutao Mou, Pei Wang, Yuanmeng Yan, Weiran Xu
Abstract
Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a taskoriented dialog system. Traditional softmaxbased confidence scores are susceptible to the overconfidence issue. In this paper, we propose a simple but strong energy-based score function to detect OOD where the energy scores of OOD samples are higher than IND samples. Further, given a small set of labeled OOD samples, we introduce an energy-based margin objective for supervised OOD detection to explicitly distinguish OOD samples from INDs. Comprehensive experiments and analysis prove our method helps disentangle confidence score distributions of IND and OOD data.- Anthology ID:
- 2022.seretod-1.5
- Volume:
- Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, Beijing (Hybrid)
- Editors:
- Zhijian Ou, Junlan Feng, Juanzi Li
- Venue:
- SereTOD
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 31–38
- Language:
- URL:
- https://aclanthology.org/2022.seretod-1.5
- DOI:
- 10.18653/v1/2022.seretod-1.5
- Cite (ACL):
- Yanan Wu, Zhiyuan Zeng, Keqing He, Yutao Mou, Pei Wang, Yuanmeng Yan, and Weiran Xu. 2022. Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning. In Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD), pages 31–38, Abu Dhabi, Beijing (Hybrid). Association for Computational Linguistics.
- Cite (Informal):
- Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning (Wu et al., SereTOD 2022)
- PDF:
- https://preview.aclanthology.org/naacl24-info/2022.seretod-1.5.pdf