Scholar
Zhiwei Steven Wu
Google Scholar ID: MbF6rTEAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
8,691
H-index
50
i10-index
111
Publications
20
Co-authors
48
list available
Publications
Resume (English only)
Academic Achievements
- Papers published:
- - Doubly-Robust LLM-as-a-Judge: Externally Valid Estimation with Imperfect Personas
- - All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning
- - Generate-then-Verify: Reconstructing Data from Limited Published Statistics
- - Orthogonal Causal Calibration
- - Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coverage
- - A Minimaximalist Approach to Reinforcement Learning from Human Feedback
- Supported by:
- - National Science Foundation (including an NSF CAREER Award)
- - Okawa Foundation
- - Open Philanthropy
- - CMU’s Block Center for Technology and Society
- - Amazon Research Award
- - Google Faculty Research Award
- - J.P. Morgan Faculty Awards
- - Meta Research awards
- - Mozilla Research Grant
- - Apple
- - Cisco Research
Research Experience
- Currently an Associate Professor in the School of Computer Science at Carnegie Mellon University, with primary appointment in the Software and Societal Systems Department (Societal Computing program), and affiliated appointments with the Machine Learning Department and the Human-Computer Interaction Institute. Also affiliated with CyLab and the Theory Group.
Education
- No specific education background information provided.
Background
- Research interests: algorithms and machine learning; particularly foundations of responsible AI (with emphasis on privacy, bias, and uncertainty considerations), interactive learning (including imitation and reinforcement learning) and its interactions with causal inference, game theory, econometrics, and language modeling.
Miscellany
- No personal interests or other information provided.
