Scholar
Nikhil Kandpal
Google Scholar ID: jGdqlOwAAAAJ
Computer Science Ph.D. Candidate, University of Toronto
Citations & Impact
Citations
2,375
H-index
8
i10-index
7
Publications
13
Co-authors
0
Contact
Publications
Resume (English only)
Academic Achievements
- - Publications:
- - AttriBoT: A Bag of Tricks for Efficiently Approximating Leave-One-Out Context Attribution (ICLR 2025)
- - User Inference Attacks Against Large Language Models (EMNLP 2024)
- - Backdoor Attacks for In-Context Learning with Language Models (ICML 2023 AdvML Workshop)
- - Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models (ICML 2023)
- - Large Language Models Struggle to Learn Long-Tail Knowledge (ICML 2023)
- - Deduplicating Training Data Mitigates Privacy Risks in Language Models (ICML 2022)
- - Music Enhancement via Image Translation and Vocoding (IEEE ICASSP 2022)
- - Universal Adversarial Triggers for Attacking and Analyzing NLP (EMNLP 2019)
Research Experience
- - Worked at Adobe Research with Oriol Nieto and Zeyu Jin on enhancing amateur music recordings
- - Worked at Google Brain with Nicholas Carlini on backdoor attacks against language models
- - Worked at Google Research with Peter Kairouz and Alina Oprea on user-level privacy attacks
- - Conducted research at the intersection of computer security and deep learning, focusing on malware detection during undergraduate studies
- - Worked as a software engineer at a proprietary trading firm after graduation
Education
- - Degree: Ph.D. candidate
- - University: University of Toronto
- - Advisor: Dr. Colin Raffel
- - Time: Fourth year
- - Undergraduate: Graduated from the University of Maryland, College Park in 2018, majoring in Computer Engineering
Background
- - Research Interests: Understanding the relationship between ML model behavior and the data it was trained on
- - Professional Field: Privacy and security, particularly of language models
Miscellany
- - Personal Interests: Not provided
Co-authors