mlsanityData to 14 Oct 2026

Data Compressibility Quantifies LLM Memorization

arXiv 2507.06056 ↗First posted 8 Jul 2025cs.CL cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) are known to memorize portions of their training data, sometimes even reproduce content verbatim when prompted appropriately. Despite substantial interest, existing LLM memorization research has offered limited insight into how training data influences memorization and largely lacks quantitative characterization. In this work, we build upon the line of research that seeks to quantify memorization through data compressibility. We analyze why prior attempts fail to yield a reliable quantitative measure and show that a surprisingly simple shift from instance-level to set-level metrics uncovers a robust phenomenon, which we term the textitEntropy--Memorization (EM) Linearity. This law states that a set-level data entropy estimator exhibits a linear correlation with memorization scores.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 43 of its 61 references that are in the map. First posted on arXiv 8 Jul 2025.

Builds on

43 of 61 references.

References · 43

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2020Scaling Laws for Neural Language Modelscited by 6,586
  3. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,411
  4. 2020Extracting Training Data from Large Language Modelscited by 2,189
  5. 2023Pythia: A Suite for Analyzing Large Language Models Across Training and Scalingcited by 1,786
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  7. 2016Understanding deep learning requires rethinking generalizationcited by 1,674
  8. 2021Membership Inference Attacks From First Principlescited by 878
  9. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 864
  10. 2022Quantifying Memorization Across Neural Language Modelscited by 763
  11. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  12. 2024OLMo: Accelerating the Science of Language Modelscited by 613
31 more
  1. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 497
  2. 2024Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Researchcited by 483
  3. 2019Does Learning Require Memorization? A Short Tale about a Long Tailcited by 460
  4. 20242 OLMo 2 Furiouscited by 340
  5. 2022Deduplicating Training Data Mitigates Privacy Risks in Language Modelscited by 314
  6. 2024LiveBench: A Challenging, Contamination-Limited LLM Benchmarkcited by 262
  7. 2023Membership Inference Attacks against Language Models via Neighbourhood Comparisoncited by 258
  8. 2022Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Modelscited by 220
  9. 2024Do Membership Inference Attacks Work on Large Language Models?cited by 195
  10. 2021Counterfactual Memorization in Neural Language Modelscited by 156
  11. 2023Emergent and Predictable Memorization in Large Language Modelscited by 148
  12. 2023ProPILE: Probing Privacy Leakage in Large Language Modelscited by 129
  13. 2021How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVENcited by 124
  14. 2024Generalization or Memorization: Data Contamination and Trustworthy Evaluation for Large Language Modelscited by 122
  15. 2021Dataset Inference: Ownership Resolution in Machine Learningcited by 116
  16. 2024LLM Dataset Inference: Did you train on my dataset?cited by 110
  17. 2022Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacycited by 100
  18. 2024Rethinking LLM Memorization through the Lens of Adversarial Compressioncited by 94
  19. 2024Generalization v.s. Memorization: Tracing Language Models' Capabilities Back to Pretraining Datacited by 85
  20. 2025How much do language models memorize?cited by 61
  21. 2024Demystifying Verbatim Memorization in Large Language Modelscited by 56
  22. 2021Membership Inference Attack Susceptibility of Clinical Language Modelscited by 52
  23. 2024Multicalibration for Confidence Scoring in LLMscited by 46
  24. 2024Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenoncited by 31
  25. 2025Extracting memorized pieces of (copyrighted) books from open-weight language modelscited by 30
  26. 2025Exploring the limits of strong membership inference attacks on large language modelscited by 22
  27. 2024Noisy Neighbors: Efficient membership inference attacks against LLMscited by 21
  28. 2023Quantifying and Analyzing Entity-level Memorization in Large Language Modelscited by 18
  29. 2024Unveiling Differences in Generative Models: A Scalable Differential Clustering Approachcited by 16
  30. 2024Measuring Non-Adversarial Reproduction of Training Data in Large Language Modelscited by 14
  31. 2023Targeted Attack on GPT-Neo for the SATML Language Model Data Extraction Challengecited by 8

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Extracting alignment data in open modelscited by 8