paper-with-me

홈 › Papers

Memorization: A Close Look at Books

2025-04-17 · Iris Ma, Ian Domingo, Alberto Krone-Martins, Pierre Baldi, Cristina V. Lopes

To what extent can entire books be extracted from LLMs? Using the Llama 3 70B family of models, and the "prefix-prompting" extraction technique, we were able to auto-regressively reconstruct, with a very high level of similarity, one entire book (Alice's Adventures in Wonderland) from just the first 500 tokens. We were also able to obtain high extraction rates on several other books, piece-wise. However, these successes do not extend uniformly to all books. We show that extraction rates of books correlate with book popularity and thus, likely duplication in the training data. We also confirm the undoing of mitigations in the instruction-tuned Llama 3.1, following recent work (Nasr et al., 2025). We further find that this undoing comes from changes to only a tiny fraction of weights concentrated primarily in the lower transformer blocks. Our results provide evidence of the limits of current regurgitation mitigation strategies and introduce a framework for studying how fine-tuning affects the retrieval of verbatim memorization in aligned LLMs.

📄 PDF Abstract BibTeX arXiv:2504.12549

Code (0)

등록된 구현이 없습니다.

Tasks

Memorization

Methods 이 논문이 사용한 방법론

LLaMA LLaMA is a collection of foundation language models ranging from 7B to 65B parameters. It is based on the transformer architecture with various improvements that were…

Similar Papers 제목 키워드 기반

Extractable Memorization From First Principles

2026-07-14 · A. Feder Cooper, Marika Swanberg, Jamie Hayes, Lea Duesterwald 외 arxiv

Recent work on extractable memorization in LLMs suffers from two contrasting validity problems. Some studies overstate extraction, e.g., relying on sequences too short to distinguish memorization from predictability. Oth…

Positional Fragility in LLMs: How Offset Effects Reshape Our Understanding of Memorization Risks

2025-05-19 · YiXuan Xu, Antoine Bosselut, Imanol Schlag

Large language models are known to memorize parts of their training data, posing risk of copyright violations. To systematically examine this risk, we pretrain language models (1B/3B/8B) from scratch on 83B tokens, mixin…

AttributeMemorization

Extracting memorized pieces of (copyrighted) books from open-weight language models

2025-05-18 · A. Feder Cooper, Aaron Gokaslan, Amy B. Cyphert, Christopher De Sa 외

Plaintiffs and defendants in copyright lawsuits over generative AI often make sweeping, opposing claims about the extent to which large language models (LLMs) have memorized plaintiffs' protected expression. Drawing on a…

Memorization

Speak, Memory: An Archaeology of Books Known to ChatGPT/GPT-4

2023-04-28 · Kent K. Chang, Mackenzie Cramer, Sandeep Soni, David Bamman

In this work, we carry out a data archaeology to infer books that are known to ChatGPT and GPT-4 using a name cloze membership inference query. We find that OpenAI models have memorized a wide collection of copyrighted m…

Memorization

Learning and Memorization

2018-07-01 · ICML 2018 7 · Satrajit Chatterjee

In the machine learning research community, it is generally believed that there is a tension between memorization and generalization. In this work we examine to what extent this tension exists by exploring if it is …

Memorization