I am a PhD student at the University of Pennsylvania where I am advised by Rajeev Alur. I previously graduated from École Polytechnique with bachelor's and master's degrees in mathematics and computer science. My current research interests revolve broadly around mechanism design and probabilistic verification for modern multi-agent systems. The applications areas that I work with include robotics and natural language reasoning.
During my undergrad and masters, I had the honor and pleasure of working with: Alessio Mansutti on optimization in integer linear-exponential programs at the IMDEA Software Institute in Madrid, Spain(SODA'26); Nathanaël Fijalkow on programmatic reinforcement learning (GenPlan'25) at the University of Warsaw; and Caterina Urban on building an abstract interpretation based tool to verify data science Jupyter notebooks(SOAP'23) at ENS Ulm, Paris. And I started my journey in the world of formal methods research working with Sergio Mover on handling communication delays arising in cyber-physical systems at École Polytechnique, Paris.
Publications
- Aaditya Naik*, Guruprerana Shabadi*, Rajeev Alur, Mayur Naik. Do We Need Frontier Models to Verify Mathematical Proofs?. Third Conference on Language Modeling (COLM), 2026.
- Guruprerana Shabadi, Rajeev Alur. Risk-Sensitive Agent Compositions. International Conference on Learning Representations (ICLR), 2026.
- S Hitarth*, Alessio Mansutti*, Guruprerana Shabadi*. Optimization Modulo Integer Linear-Exponential Programs. ACM-SIAM Symposium on Discrete Algorithms (SODA), 2026.
Preprints
- Guruprerana Shabadi, Kaushik Mallik. Auction-Based Online Policy Adaptation for Evolving Objectives. Preprint.
Workshop Papers
- Guruprerana Shabadi, Nathanaël Fijalkow, Théo Matricon. Programmatic Reinforcement Learning: Navigating Gridworlds. Generalization in Planning Workshop, AAAI 2025.
- Luca Negrini, Guruprerana Shabadi, Caterina Urban. Static Analysis of Data Transformations in Jupyter Notebooks. SOAP, ACM SIGPLAN Conference on Programming Language Design and Implementation, 2023.