Extracting memorized pieces of (copyrighted) books from open-weight language models
A. Feder Cooper · Mark Lemley · Allison Casasola · Ahmed M Ahmed · Aaron Gokaslan · Amy B. Cyphert · Christopher De Sa · Daniel E. Ho · Percy Liang
We develop new methods and visualizations for surfacing memorization of (copyrighted) books in open-weight LLMs, revealing far more memorization than previously understood
摘要
Plaintiffs and defendants in copyright lawsuits make sweeping, opposing claims about the extent to which large language models (LLMs) memorize protected expression from books in their training data. We show that these polarized positions dramatically oversimplify the relationship between memorization and copyright. To do so, we develop a technique to measure memorization of books, which we apply to 200 books and 14 open-weight LLMs. Through over 3000 experiments, we show that memorization varies both by model and book. While we find that most LLMs do not memorize most books either in whole or in part, there are notable exceptions. Llama 3.1 70B entirely memorizes some books, like Harry Potter and the Sorcerer's Stone; memorization is so extensive that one can deterministically extract the whole book almost verbatim using the book's first few words as an initial prompt. We discuss why our results have significant implications for copyright cases, though not ones that unambiguously favor either side.