(2023.05.23) Accepted at TrustNLP 2023

A survey paper has been accepted at the TrustNLP workshop at ACL2023. The paper reviewed literature on the memorization properties and training data extraction of Large Language Models, and discussed future directions.

Shotaro Ishihara (2023). Training Data Extraction From Pre-trained Language Models: A Survey. Proceedings of Third Workshop on Trustworthy Natural Language Processing. [ arXiv ]