USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation
Article Status
Published
Authors/contributors
- Mehri, Shikib (Author)
- Eskenazi, Maxine (Author)
Title
USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation
Date
2020
Proceedings Title
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Conference Name
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Place
Online
Publisher
Association for Computational Linguistics
Pages
681-707
Language
en
Short Title
USR
Accessed
27/10/2023, 17:30
Library Catalogue
DOI.org (Crossref)
Extra
<AI Smry>: USR is a reference-free metric that trains unsupervised models to measure several desirable qualities of dialog and is shown to strongly correlate with human judgment on both Topical-Chat and PersonaChat.
Citation
Mehri, S., & Eskenazi, M. (2020). USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 681–707. https://doi.org/10.18653/v1/2020.acl-main.64
Technical methods
Link to this record