About me
I am a Ph.D. student @ Kahlert School of Computing, University of Utah. I am advised by Dr. Guanhong Tao at the SaLT Lab, broadly focusing on ML/AI security and robustness. I am currently exploring the various vulnerabilities and unexpected behaviors of generative AI systems such as LLMs and T2I difusion models. My primary motivation is to uncover vulnerabilities and aid in the development of models that are not only robust, secure, and transparent, but also efficiently designed.
I received my Master of Engineering Science and my B.S. in Computer Science from the University of Mississippi, under the supervision of Dr. Charles Walter, where I focused on exploiting the vulnerabilities and of the state-of-the-art object detection models to develop generalizable adversarial attacks.
Publications
Generalized Loss-Function-Based Attacks for Object Detection Models - Soumil Datta, and Charles Walter. Proceedings of the 58th Hawaii International Conference on System Sciences (HICSS), 2025 (Nominated for Best Paper)
EllipScape: A Genetic Algorithm Based Approach to Non-Photorealistic Colored Image Reconstruction for Evolutionary Art - Soumil Datta, and Charles Walter. Proceedings of the 57th Hawaii International Conference on System Sciences (HICSS), 2024 (Best Paper Award, Top 1%)
Awards & Honors
- HICSS 2024 Conference Best Paper Award, Jan 2024
- Marcus Elvis Taylor Memorial Medal, April 2022
- Outstanding Student in Computer Science Award, Mar 2021