Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?

Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise? Xinzhe Li author Ming Liu author Shang Gao author 2023-07 text Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023) Alexis Palmer editor Jose Camacho-collados editor Association for Computational Linguistics Toronto, Canada conference publication li-etal-2023-pretrained 10.18653/v1/2023.starsem-1.15 https://aclanthology.org/2023.starsem-1.15/ 2023-07 165 173