Adjunct Professor
Shrimai Prabhumoye is an Adjunct Professor at Boston University’s Department of Computer Science and a Senior Research Scientist with the Applied Deep Learning Research Group at Nvidia. Her research is dedicated to advancing the state-of-the-art in large language models (LLMs) by enhancing their reasoning capabilities and ensuring their safety through rigorous mitigation of toxicity and bias. As the lead contributor to the Nemotron family of models, she has worked extensively on data curation, pretraining, and scaling. Her current work focuses on optimizing pretraining pipelines with an emphasis on data selection, blending, and ordering strategies to maximize downstream model accuracy. She is particularly focused on improving reasoning in LLMs, including generating synthetic data for advanced mathematical reasoning and enabling models to handle longer, more complex reasoning tasks that require deeper thought and understanding. Her work has featured in many media outlets like VentureBeat, Forbes and TechCrunch.
Previously, she graduated with a PhD from School of Computer Science, Carnegie Mellon University. At CMU, she was fortunate to be advised by Prof. Alan W. Black and Prof. Ruslan Salakhutdinov. Her thesis focused on controllable text generation with a focus on style, content and structure, as well as its ethical considerations. She co-designed the Computational Ethics for NLP course which was offered for the first time in Spring 2018 at CMU. She graduated with a Masters in Language Technologies in Aug 2017. During that time, she was leading the CMU Magnus team in the Amazon Alexa Prize competition. She completed my undergraduate at National Institute of Technology, Karnataka, India.