Nicolas Papernot
Associate Professor
Studies security and privacy of machine learning, including differentially private learning and machine unlearning.
Nicolas Papernot's lab at Toronto works on trustworthy & safe ML. 26 papers since 2023, 19 at top venues, mostly ICML and FAccT. 7 PhD students and 4 master's students. 43 alumni, most often to Anthropic and Cohere. Applications close .
At a glance
- papers, 2023–2026
- 7/ year
- at T1 venues
- 73%
- PhD students
- 7
- MS students
- 4
- alumni listed
- 43
Papers by yearfrom the lab homepage, 2023–2026
- 8 papers2026 · T1 6 · collecting
- 17 papers2025 · T1 13
- 1 papers2024 · T1 0
- 0 papers2023 · T1 0
T1 venuesOtherVenues: ICML, FAccT, ICLR, NeurIPS, TMLR
Research focuspaper titles that use each topic's words
- Trustworthy & safe ML3 papers
- machine unlearning3 papers
- differentially1 paper
- differentially private1 paper
- studies security0 papers
Paper list
From the lab homepage.
20268 papers
- NeurIPS (T1)CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents (opens in a new tab)
- USENIX Security (T1)Certified in Theory, Broken in Practice: Assumption Gaps in Cryptographic Model Certification (opens in a new tab)
- ICML (T1)Efficient Public Verification of Private ML via Regularization (opens in a new tab)
- Proceedings of the IEEE Conference on Secure and Trustworthy Machine LearningGauss-Newton Unlearning for the LLM Era (opens in a new tab)
- ICML (T1)OptiFluence: Principled Design of Privacy Canaries
- NDSS (T1)Preempt: Sanitizing Sensitive Prompts for LLMs (opens in a new tab)
- ICLR (T1)Private Rate-Constrained Optimization with Applications to Fair Learning (opens in a new tab)
- FAccTSocietal Alignment Frameworks Can Improve LLM Alignment (opens in a new tab)
202517 papers
- ICLR (T1)A False Sense of Safety: Unsafe Information Leakage in Safe AI Responses (opens in a new tab)
- S&P (T1)Architectural Neural Backdoors from First Principles (opens in a new tab)
- Proceedings of the IEEE Conference on Secure and Trustworthy Machine LearningBackdoor Detection through Duplicated Execution of Outsourced Training (opens in a new tab)
- ICML (T1)Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention (opens in a new tab)
- ICML (T1)Fast Exact Unlearning of Fine-Tuning Data for LLMs (opens in a new tab)
- Proceedings of the IEEE Conference on Secure and Trustworthy Machine LearningInexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy (opens in a new tab)
- ICML (T1)Language Models May Verbatim Complete Text They Were Not Explicitly Trained On (opens in a new tab)
- TMLR (T1)Learned-Database Systems Security (opens in a new tab)
- ICML (T1)Leveraging Per-Instance Privacy for Machine Unlearning
- NeurIPS (T1)Position Paper: Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy, Research, and Practice (opens in a new tab)
- CCS (T1)Secure Noise Sampling for Differentially Private Collaborative Learning (opens in a new tab)
- TMLR (T1)Selective Prediction via Training Dynamics (opens in a new tab)
Where alumni went43 listed, 41% to industry
Industry 16Academia 23
Most often to
- Anthropic
- Cohere
- Google DeepMind
Also to
- University of Toronto
- Stanford University
- Cornell University
Apply to TorontoFall 2027
Computer Science MSc/PhD
- Round
- Single round (Dec 1)
- Opens
- —
- GRE
- optional
- English
- TOEFL/IELTS if applicable
- Fee
- CAD 135
Confirmed on the program page.
Similar labsnearest by research summary
- USUW
Simon S. Du
- Learning theory & optimization
- GBOxford
Bernardo Cuenca Grau
- Reasoning & neuro-symbolic AI
- USUIUC
Olgica Milenkovic
- Learning theory & optimization
- USUSC
Haipeng Luo
- Learning theory & optimization
- SGNUS
Mehul Motani
- Signal processing & information theory
- GBOxford
Nicholas Teh
- Trustworthy & safe ML
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Torontocloses