Published multiple papers at top venues including ICML 2025, AAAI 2025, IJCAI 2024, AAAI 2024, ECML-PKDD 2023, JMLR 2022, and TheWebConf 2022
Proposed embedding Automated Reasoning (AR) solvers as differentiable layers in neural networks to guarantee constraint satisfaction in structured prediction tasks
Developed approaches that successfully discover ground-truth scientific expressions with up to 50 variables, significantly outperforming prior methods (limited to ~3 variables)
PhD Thesis: 'Integrating Automated Reasoning with Machine Learning for Structured Prediction and Scientific Discovery'
Contributed a book chapter to Springer’s 'Computational Approaches to Scientific Discovery' (2025)