paper-with-me

Papers

Undesirable Memorization in Large Language Models: A Survey

2024-10-03 · Ali Satvaty, Suzan Verberne, Fatih Turkmen

While recent research increasingly showcases the remarkable capabilities of Large Language Models (LLMs), it's vital to confront their hidden pitfalls. Among these challenges, the issue of memorization stands out, posing significant ethical and legal risks. In this paper, we presents a Systematization of Knowledge (SoK) on the topic of memorization in LLMs. Memorization is the effect that a model tends to store and reproduce phrases or passages from the training data and has been shown to be the fundamental issue to various privacy and security attacks against LLMs. We begin by providing an overview of the literature on the memorization, exploring it across five key dimensions: intentionality, degree, retrievability, abstraction, and transparency. Next, we discuss the metrics and methods used to measure memorization, followed by an analysis of the factors that contribute to memorization phenomenon. We then examine how memorization manifests itself in specific model architectures and explore strategies for mitigating these effects. We conclude our overview by identifying potential research topics for the near future: to develop methods for balancing performance and privacy in LLMs, and the analysis of memorization in specific contexts, including conversational agents, retrieval-augmented generation, multilingual language models, and diffusion language models.

📄 PDF Abstract BibTeX arXiv:2410.02650

Code (0)

등록된 구현이 없습니다.

Tasks

MemorizationRetrieval-augmented GenerationSurvey

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Emergent and Predictable Memorization in Large Language Models

2023-04-21 · NeurIPS 2023 11 · Stella Biderman, USVSN Sai Prashanth, Lintang Sutawika, Hailey Schoelkopf 외

Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a …

Memorization

Rethinking Memorization Measures and their Implications in Large Language Models

2025-07-20 · Bishwamittra Ghosh, Soumi Das, Qinyuan Wu, Mohammad Aflah Khan 외 arxiv

Concerned with privacy threats, memorization in LLMs is often seen as undesirable, specifically for learning. In this paper, we study whether memorization can be avoided when optimally learning a language, and whether th…

Quantifying Memorization Across Neural Language Models

2022-02-15 · Nicholas Carlini, Daphne Ippolito, Matthew Jagielski, Katherine Lee 외

Large language models (LMs) have been shown to memorize parts of their training data, and when prompted appropriately, they will emit the memorized training data verbatim. This is undesirable because memorization violate…

FairnessMemorization

Do LLMs Memorize Recommendation Datasets? A Preliminary Study on MovieLens-1M

2025-05-15 · Dario Di Palma, Felice Antonio Merra, Maurizio Sfilio, Vito Walter Anelli 외

Large Language Models (LLMs) have become increasingly central to recommendation scenarios due to their remarkable natural language understanding and generation capabilities. Although significant research has explored the…

BenchmarkingMemorizationNatural Language UnderstandingRecommendation Systems

How BPE Affects Memorization in Transformers

2021-10-06 · Eugene Kharitonov, Marco Baroni, Dieuwke Hupkes

Training data memorization in NLP can both be beneficial (e.g., closed-book QA) and undesirable (personal data extraction). In any case, successful model training requires a non-trivial amount of memorization to store wo…

Memorization