2023
Emergent and Predictable Memorization in Large Language Models
Stella Biderman, USVSN Sai Prashanth, Lintang Sutawika, Hailey Schoelkopf, Quentin Gregory Anthony, Shivanshu Purohit +1
NeurIPS 2023poster
Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for deploying language models. In particular, it is vital to minimize a model's memorization of sensitive datapoints such as those containing personal identifi…