NP
Nicolas Papernot
cs.LGcs.CRstat.MLcs.AIcs.CVcs.CYcs.CLcs.NEcs.SDeess.AS
On Valency
published · living versionsW_ez3a9a2w·v1 · currentpublished
Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
with Patrick McDaniel, Ian Goodfellow
1 version
Preprints & journals
131 papers in the corpus · 2015–2026Empirical Analysis of Evasion and Poisoning Against Malware Data Drift Detection2608.03642v3 · Mingyue Yang, David Lie, Nicolas Papernot2026 · 0 citationsarXiv
Certified in Theory, Broken in Practice: Assumption Gaps in Cryptographic Model Certification2607.21839v1 · Carter Luck, Olive Franzese-McLaughlin, Elisaweta Masserova et al.2026 · 0 citationsarXiv
Private Rate-Constrained Optimization with Applications to Fair Learning2505.22703v2 · Mohammad Yaghini, Tudor Cebere, Michael Menart et al.2025 · 0 citationsarXiv
Efficient Public Verification of Private ML via Regularization2512.04008v2 · Zoe Ruha Bell, Anvith Thudi, Olive Franzese-McLaughlin et al.2025 · 0 citationsarXiv
Societal Alignment Frameworks Can Improve LLM Alignment2503.00069v2 · Karolina Sta'nczak, Nicholas Meade, Mehar Bhatia et al.2025 · 0 citationsarXiv
CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents2601.09923v3 · Hanna Foerster, Tom Blanchard, Kristina Nikoli'c et al.2026 · 0 citationsarXiv
AI Agents Enable Adaptive Computer Worms2606.03811v1 · Jonas Guan, Tom Blanchard, Hanna Foerster et al.2026 · 2 citationsarXiv
Beyond Laplace and Gaussian: Exploring the Generalized Gaussian Mechanism for Private Machine Learning2506.12553v2 · Roy Rinberg, Ilia Shumailov, Vikrant Singhal et al.2025 · 0 citationsarXiv
Gauss-Newton Unlearning for the LLM Era2602.10568v1 · Lev McKinney, Anvith Thudi, Juhan Bae et al.2026 · 0 citationsarXiv
Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research2412.06966v2 · A. Feder Cooper, Christopher A. Choquette-Choo, Miranda Bogen et al.2024 · 3 citationsarXiv
What Does It Take to Build a Performant Selective Classifier?2510.20242v2 · Stephan Rabanser, Nicolas Papernot2025 · 2 citationsarXiv
When Vision Fails: Text Attacks Against ViT and OCR2306.07033v2 · Nicholas Boucher, Jenny Blessing, Ilia Shumailov et al.2023 · 0 citationsarXiv
Fast Exact Unlearning for In-Context Learning Data for LLMs2402.00751v2 · Andrei I. Muresanu, Anvith Thudi, Michael R. Zhang et al.2024 · 1 citationarXiv
Learned-Database Systems Security2212.10318v4 · Roei Schuster, Jin Peng Zhou, Thorsten Eisenhofer et al.2022 · 0 citationsarXiv
Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention2505.23968v1 · Stephan Rabanser, Ali Shahin Shamsabadi, Olive Franzese et al.2025 · 0 citationsarXiv
Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model2405.14457v3 · Tudor Cebere, Aur'elien Bellet, Nicolas Papernot2024 · 1 citationarXiv
Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings2505.22356v1 · Ang'eline Pouget, Mohammad Yaghini, Stephan Rabanser et al.2025 · 0 citationsarXiv
Leveraging Per-Instance Privacy for Machine Unlearning2505.18786v1 · Nazanin Mohammadi Sepahvand, Anvith Thudi, Berivan Isik et al.2025 · 0 citationsarXiv
Author Correction: AI models collapse when trained on recursively generated data.40119087 · Shumailov, Ilia, Shumaylov, Zakhar, Zhao, Yiren et al.2025 · 4 citationsNature. 2025;640(8058):E6
Backdoor Detection through Replicated Execution of Outsourced Training2504.00170v1 · Hengrui Jia, Sierra Wyllie, Akram Bin Sediq et al.2025 · 0 citationsarXiv
Verifiable and Provably Secure Machine Unlearning2210.09126v3 · Thorsten Eisenhofer, Doreen Riepel, Varun Chandrasekaran et al.2022 · 22 citationsarXiv
Augment then Smooth: Reconciling Differential Privacy with Certified Robustness2306.08656v3 · Jiapeng Wu, Atiyeh Ashari Ghomi, David Glukhov et al.2023 · 3 citationsarXiv
On the Privacy Risk of In-context Learning2411.10512v1 · Haonan Duan, Adam Dziedzic, Mohammad Yaghini et al.2024 · 2 citationsarXiv
Temporal-Difference Learning Using Distributed Error Signals2411.03604v1 · Jonas Guan, Shon Eduard Verch, Claas Voelcker et al.2024 · 1 citationarXiv
Breach By A Thousand Leaks: Unsafe Information Leakage in `Safe' AI Responses2407.02551v2 · David Glukhov, Ziwen Han, Ilia Shumailov et al.2024 · 0 citationsarXiv
AI models collapse when trained on recursively generated data.39048682 · Shumailov, Ilia, Shumaylov, Zakhar, Zhao, Yiren et al.2024 · 749 citationsNature. 2024;631(8022):755-759
Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD2307.00310v3 · Anvith Thudi, Hengrui Jia, Casey Meehan et al.2023 · 0 citationsarXiv
UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI2407.00106v1 · Ilia Shumailov, Jamie Hayes, Eleni Triantafillou et al.2024 · 1 citationarXiv
Memorization in Self-Supervised Learning Improves Downstream Generalization2401.12233v3 · Wenhao Wang, Muhammad Ahmad Kaleem, Adam Dziedzic et al.2024 · 3 citationsarXiv
Beyond Labeling Oracles: What does it mean to steal ML models?2310.01959v3 · Avital Shafran, Ilia Shumailov, Murat A. Erdogdu et al.2023 · 0 citationsarXiv
LLM Dataset Inference: Did you train on my dataset?2406.06443v1 · Pratyush Maini, Hengrui Jia, Nicolas Papernot et al.2024 · 15 citationsarXiv
Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy2403.01218v3 · Jamie Hayes, Ilia Shumailov, Eleni Triantafillou et al.2024 · 12 citationsarXiv
Decentralised, Collaborative, and Privacy-preserving Machine Learning for Multi-Hospital Data2402.00205v2 · Congyu Fang, Adam Dziedzic, Lin Zhang et al.2024 · 34 citationseBioMedicine, vol. 101, p. 105006, 2024
The Curse of Recursion: Training on Generated Data Makes Models Forget2305.17493v3 · Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao et al.2023 · 159 citationsarXiv
Advancing Differential Privacy: Where We Are Now and Future Directions for Real-World Deployment2304.06929v2 · Rachel Cummings, Damien Desfontaines, David Evans et al.2023 · 49 citationsarXiv
Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias2403.07857v1 · Sierra Wyllie, Ilia Shumailov, Nicolas Papernot2024 · 28 citationsarXiv
Auditing Private Prediction2402.09403v1 · Karan Chadha, Matthew Jagielski, Nicolas Papernot et al.2024 · 1 citationarXiv
Architectural Neural Backdoors from First Principles2402.06957v1 · Harry Langford, Ilia Shumailov, Yiren Zhao et al.2024 · 5 citationsarXiv
Regulation Games for Trustworthy Machine Learning2402.03540v1 · Mohammad Yaghini, Patty Liu, Franziska Boenisch et al.2024 · 2 citationsarXiv
Have it your way: Individualized Privacy Assignment for DP-SGD2303.17046v2 · Franziska Boenisch, Christopher Muhl, Adam Dziedzic et al.2023 · 10 citationsarXiv
Author Correction: Subtle adversarial image manipulations influence both human and machine perception.38228579 · Veerabadran, Vijay, Goldman, Josh, Shankar, Shreya et al.2024 · 0 citationsNature communications. 2024;15(1):543
Robust and Actively Secure Serverless Collaborative Learning2310.16678v1 · Olive Franzese, Adam Dziedzic, Christopher A. Choquette-Choo et al.2023 · 2 citationsarXiv
The Adversarial Implications of Variable-Time Inference2309.02159v1 · Dudi Biton, Aditi Misra, Efrat Levy et al.2023 · 0 citationsarXiv
Subtle adversarial image manipulations influence both human and machine perception.37582834 · Veerabadran, Vijay, Goldman, Josh, Shankar, Shreya et al.2023 · 16 citationsNature communications. 2023;14(1):4933
LLM Censorship: A Machine Learning Challenge or a Computer Security Problem?2307.10719v1 · David Glukhov, Ilia Shumailov, Yarin Gal et al.2023 · 19 citationsarXiv
Training Private Models That Know What They Don't Know2305.18393v1 · Stephan Rabanser, Anvith Thudi, Abhradeep Thakurta et al.2023 · 0 citationsarXiv
Tubes Among Us: Analog Attack on Automatic Speaker Identification2202.02751v2 · Shimaa Ahmed, Yash Wani, Ali Shahin Shamsabadi et al.2022 · 1 citationarXiv
Flocks of Stochastic Parrots: Differentially Private Prompt Learning for Large Language Models2305.15594v1 · Haonan Duan, Adam Dziedzic, Nicolas Papernot et al.2023 · 12 citationsarXiv
Measuring Forgetting of Memorized Training Examples2207.00099v2 · Matthew Jagielski, Om Thakkar, Florian Tramer et al.2022 · 15 citationsarXiv
Proof-of-Learning is Currently More Broken Than You Think2208.03567v2 · Congyu Fang, Hengrui Jia, Anvith Thudi et al.2022 · 19 citationsarXiv
Career total: 222 works. 131 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.