Debangshu Banerjee

CS PhD Student @ UIUC

I am a 3rd year PhD student in the Computer Science department at the University of Illinois, Urbana-Champaign. I work at the intersection of machine learning and formal methods to develop next-generation programming systems with provable correctness. I am presently advised by Prof. Gagandeep Singh.

I am currently looking into the development of programming systems with LLMs that provide users with finer-grained control over the generated output, such as code and logical formulas. On the theoretical side, I am interested in investigating the capabilities and potential limitations of LLMs in understanding code semantics. Previously, my research focused on neural network verification and certifiable training, where I developed the first scalable GPU-accelerated verification and training method for relational properties, including monotonicity and robustness against universal adversarial perturbations (UAPs).

Selected Publications

  1. CRANE
    CRANE: Expressive Grammar-Constrained LLM Generation
    Banerjee, Debangshu, Suresh, Tarun, Ugare, Shubham, Misailovic, Sasa, and Singh, Gagandeep
    In Preprint 2025
  2. CIVET
    Support is All You Need for Certified VAE Training
    Xu, Calvin,  Banerjee, Debangshu, Vasisht, Deepak, and Singh, Gagandeep
    In The Thirteenth International Conference on Learning Representations (ICLR) 2025
  3. RABBit
    Relational Verification Leaps Forward with RABBit
    Suresh, Tarun,  Banerjee, Debangshu, and Singh, Gagandeep
    In The Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS) 2024
  4. RACoon
    Relational DNN Verification With Cross Executional Bound Refinement
    Banerjee, Debangshu, and Singh, Gagandeep
    In Forty-first International Conference on Machine Learning (ICML) 2024
  5. RaVeN
    Input-Relational Verification of Deep Neural Networks
    Banerjee, Debangshu, Xu, Calvin, and Singh, Gagandeep
    In Programming Language Design and Implementation (PLDI) 2024
  6. ProFIt
    Interpreting Robustness Proofs of Deep Neural Networks
    Banerjee, Debangshu, Singh, Avaljot, and Singh, Gagandeep
    In The Twelfth International Conference on Learning Representations (ICLR) 2024
  7. IRS
    Incremental Randomized Smoothing Certification
    Ugare, Shubham, Suresh, Tarun,  Banerjee, Debangshu, Singh, Gagandeep, and Misailovic, Sasa
    In The Twelfth International Conference on Learning Representations (ICLR) 2024
  8. IVAN
    Incremental Verification of Neural Networks
    Ugare, Shubham,  Banerjee, Debangshu, Misailovic, Sasa, and Singh, Gagandeep
    Programming Language Design and Implementation (PLDI) Jun 2023
Affiliations & Internships
               
IIT Guwahati
2016-2020
B. Tech Computer Science
Google
Summer 2019
SDE Intern
Google
2020-2022
Software Engineer (L4)
UIUC
2022-Present
PhD Computer Science