An Empirical Study on Pseudo-log-likelihood Bias Measures for Masked Language Models Using Paraphrased Sentences

An Empirical Study on Pseudo-log-likelihood Bias Measures for Masked Language Models Using Paraphrased Sentences Bum Chul Kwon author Nandana Mihindukulasooriya author 2022-07 text Proceedings of the 2nd Workshop on Trustworthy Natural Language Processing (TrustNLP 2022) Apurv Verma editor Yada Pruksachatkun editor Kai-Wei Chang editor Aram Galstyan editor Jwala Dhamala editor Yang Trista Cao editor Association for Computational Linguistics Seattle, U.S.A. conference publication kwon-mihindukulasooriya-2022-empirical 10.18653/v1/2022.trustnlp-1.7 https://aclanthology.org/2022.trustnlp-1.7/ 2022-07 74 79