David Acuna

I am a Senior Research Scientist at NVIDIA Research in the Language and Cognition Research Group working with Yejin Choi. Prior to this, I was the first full-time employee of the NV Toronto AI Lab led by Sanja Fidler. I obtained my PhD in Machine Learning and Computer Vision from the University of Toronto, where I was advised by Prof. Sanja Fidler. During my PhD, I was also affiliated with the Vector Institute for AI. In 2018, I completed my Master’s Degree in Applied Computing at the same institution. I am honored to have received the 2020 Microsoft Ada Lovelace Fellowship. My current research focuses on reasoning models, with an emphasis on synthetic data, inference-time scaling, agents, RL, and the role of data synthesis and reasoning in the path toward AGI and Physical AI. I am also particularly interested in exploring optimal strategies for adapting foundation models to deliver enterprise value. In the past, I have worked on synthetic data generation and domain adaptation for visual learning. More broadly, my research interests span representation learning, model adaptation, controllable generation, synthetic data, optimization, and generative modeling. I also have a long-standing interest in scene understanding and low-level vision. Graduate students interested in internships at NVIDIA are welcome to contact me with a CV and summary of research interests. More info: http://www.cs.toronto.edu/~davidj/