Sepideh Mamooler

I’m a PhD student in Computer Science at the Swiss Federal Institute of Technology Lausanne (EPFL), co-advised by Prof. Antoine Bosselut (NLP) and Prof. Alexander Mathis (Computational Neuroscience). I am a Boehringer Ingelheim Fonds PhD Fellow and a former intern at Google DeepMind, where I studied social reasoning in large language models under the supervision of Florian Hartmann and Iulia Comșa.

My research addresses a critical gap in multimodal AI: current foundation models lack principled mechanisms for expressing and handling uncertainty. I develop training and evaluation frameworks that make these models reliably calibrated when inputs are ambiguous, noisy, or lacking critical information.