Scholar
Debangshu Banerjee
Google Scholar ID: G5dhKqAAAAAJ
PhD Student, Computer Science, UIUC
Citations & Impact
Citations
109
H-index
6
i10-index
5
Publications
16
Co-authors
9
list available
Publications
Resume (English only)
Background
- 4th-year PhD student in Computer Science at the University of Illinois, Urbana-Champaign (UIUC)
- Works at the intersection of machine learning and formal methods to develop next-generation programming systems with provable correctness
- Currently advised by Prof. Gagandeep Singh
- PhD supported by the Bloomberg PhD Fellowship
- Research focuses on giving users direct control over LLM outputs, including enforcing safety and access control in generated agents; also investigates theoretical capabilities and limitations of LLMs in understanding code semantics
- Previously worked on neural network verification and certifiable training, developing the first scalable GPU-accelerated method for verifying relational properties such as monotonicity and robustness against universal adversarial perturbations (UAPs)
