Nathalie Baracaldo

IBM (United States) — United States

Computer Science · Physical Sciences

26h-index3.4kcitations98works2.42yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Privacy-Preserving Technologies in Data(50), Cryptography and Data Security(26), Stochastic Gradient Optimization Techniques(16), Adversarial Robustness in Machine Learning(16), Privacy, Security, and Data Protection(10)

Publications98 total

In-Context Probing for Membership Inference in Fine-Tuned Language Models
arXiv (Cornell University)·2025Open Access
Beyond SFT: Reinforcement Learning for Safer Large Reasoning Models with Better Reasoning Ability
arXiv (Cornell University)·2025Open Access
Membership Inference Attacks as Privacy Tools: Reliability, Disparity and Ensemble
2025Open Access
Mitigating Modality Imbalance in Multi-modal Learning via Multi-objective Optimization
arXiv (Cornell University)·2025Open Access
FedKDD 2025: The 2025 International Joint Workshop on Federated Learning for Data Mining and Graph Analytics
2025· 1 citedOpen Access
Evaluating the Dynamics of Membership Privacy in Deep Learning
arXiv (Cornell University)·2025Open Access
Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills
arXiv (Cornell University)·2025Open Access
Invariance Makes LLM Unlearning Resilient Even to Unanticipated Downstream Fine-Tuning
arXiv (Cornell University)·2025Open Access
Rethinking machine unlearning for large language models
UNC Libraries·2025Open Access
Rethinking machine unlearning for large language models
Nature Machine Intelligence·2025· 46 cited
Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills
2025Open Access
Towards a Re-evaluation of Data Forging Attacks in Practice
arXiv (Cornell University)·2024Open Access
MAP: Multi-Human-Value Alignment Palette
arXiv (Cornell University)·2024Open Access
WAGLE: Strategic Weight Attribution for Effective and Modular Unlearning in Large Language Models
arXiv (Cornell University)·2024· 1 citedOpen Access
Understanding Forgetting in LLM Supervised Fine-Tuning and Preference Learning -- A Convex Optimization Perspective
arXiv (Cornell University)·2024· 1 citedOpen Access
FedKDD: International Joint Workshop on Federated Learning for Data Mining and Graph Analytics
2024· 1 cited
Turning Generative Models Degenerate: The Power of Data Poisoning Attacks
arXiv (Cornell University)·2024Open Access
Byzantine-Resilient Bilevel Federated Learning
2024· 1 cited
Split, Unlearn, Merge: Leveraging Data Attributes for More Effective Unlearning in LLMs
arXiv (Cornell University)·2024Open Access
Rethinking Machine Unlearning for Large Language Models
arXiv (Cornell University)·2024· 4 citedOpen Access
page 1 of 5Next →

Frequent Co-authors

Heiko Ludwig(27), Yi Zhou(19), Ali Anwar(16), James Joshi(8), Runhua Xu(6), Swanand Kadhe(5), Stacey Truex(4), Bryant Chen(4), Thomas Steinke(3), Rui Zhang(3), Ahsan Ali(3), Syed Zawad(3), Yue Cheng(3), Taesung Lee(3), Yuya Jeremy Ong(3), Mathieu Sinn(3), Ambrish Rawat(3), Annie Abay(3), Zheng Chai(2), Feng Yan(2)