Manushree Vasu

I’m a PhD student in Computer Science at Boston University, advised by Prof. Deepti Ghadiyaram, working on representation learning, generative models, and multimodal/vision-language models.

I completed my MS in Computer Science at Georgia Tech, advised by Prof. Humphrey Shi and closely working with Prof. Judy Hoffman, where I built generalist multimodal LLMs spanning image, audio, and video, and worked on improving the smoothness of diffusion model latent spaces (Smooth Diffusion, CVPR 2024). During my MS, I was also an ML Research intern with Adobe Research during summer 2024, where I worked at the intersection of HCI and NLP under Balaji Krishnamurthy.

I previously interned at the Artificial Intelligence and Robotics Lab at IISc Bangalore, advised by Prof. Suresh Sundaram and Dr. Chandan Gautam on zero-shot learning and domain generalization. I also interned at IBM Research under Dr. Diptikalyan Saha on interpretability repair of ML models, and at Neural Garage building the voice-cloning module for VisualDub. I also had the privilege of working with Prof. Harish Kumar JR.

On the side, I also mentor undergraduate students, getting started with research at the Research Society Manipal.