Scholar
Shengyuan Hu
Google Scholar ID: m_ZHHToAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
1,948
H-index
11
i10-index
11
Publications
17
Co-authors
9
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers such as 'Position: LLM Unlearning Benchmarks are Weak Measures of Progress', 'Jogging the Memory of Unlearned Model Through Targeted Relearning Attack', and received Best Paper Award at ICLR 2022 Workshop on Socially Responsible Machine Learning (SRML) and ICLR 2021 Secure ML Workshop.
Research Experience
- Research Intern at FAIR (2023), Research Intern at Meta Platforms, Inc. (2021), Research Engineer Intern at Uber Technologies, Inc. (2019), Software Engineer Intern at MasterClass (2018). Also served as a Teaching Assistant at Cornell and CMU for various courses.
Education
- PhD: Machine Learning Department at Carnegie Mellon University, Advisor: Virginia Smith; BA: Cornell University, Double Major in Computer Science and Mathematics
Background
- Research interests include LLM alignment, machine unlearning, differential privacy, and trustworthy federated learning methods. Focuses on studying analytical formulations and efficient methods that provide both theoretical guarantees and strong empirical performance on real-world applications to handle these issues in federated learning.
Miscellany
- Reviewer for ICML (2021-2024), ICLR (2021-2024), NeurIPS (2021-2024), TMLR; PC Member for NeurIPS 2022 Workshop on Trustworthy and Socially Responsible Machine Learning (TSRML), FL4NLP@ACL 2022 Workshop; Reviewer for IEEE Network Magazine
