PPaperPicks

Matthew Jagielski

12 papers at tracked venues · 11 at CORE A* · active 20242025

Venues

Frequent coauthors

Papers

  1. Differentially Private Prototypes for Imbalanced Transfer Learning
  2. Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards
  3. Exploring the limits of strong membership inference attacks on large language models
  4. Language Models May Verbatim Complete Text They Were Not Explicitly Trained On
  5. Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research
  6. On Evaluating the Durability of Safeguards for Open-Weight LLMs
  7. Privacy Ripple Effects from Adding or Removing Personal Information in Language Model Training
  8. Scalable Extraction of Training Data from Aligned, Production Language Models
  9. The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD
  10. Auditing Private Prediction
  11. Stealing part of a production language model
  12. Synthetic Query Generation for Privacy-Preserving Deep Retrieval Systems using Differentially Private Language Models