Using Membership Inference Attacks to Evaluate Privacy-Preserving Language Modeling Fails for Pseudonymizing Data

Using Membership Inference Attacks to Evaluate Privacy-Preserving Language Modeling Fails for Pseudonymizing Data Thomas Vakili author Hercules Dalianis author 2023-05 text Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa) Tanel Alumäe editor Mark Fishel editor University of Tartu Library Tórshavn, Faroe Islands conference publication vakili-dalianis-2023-using https://aclanthology.org/2023.nodalida-1.33/ 2023-05 318 323