You Are What You Write: Author re-identification privacy attacks in the era of pre-trained language models
(2024)
Journal Article
Plant, R., Giuffrida, V., & Gkatzia, D. (online). You Are What You Write: Author re-identification privacy attacks in the era of pre-trained language models. Computer Speech and Language, https://doi.org/10.1016/j.csl.2024.101746
The widespread use of pre-trained language models has revolutionised knowledge transfer in natural language processing tasks. However, there is a concern regarding potential breaches of user trust due to the risk of re-identification attacks, where m... Read More about You Are What You Write: Author re-identification privacy attacks in the era of pre-trained language models.