Jahnavi Yelamanchi

Hi! I’m a computer engineering graduate student and a Graduate Researcher at the General Purpose Robotics and AI Lab (GRAIL) at NYU’s CILVR Group with Prof. Lerrel Pinto.

Other experiences of mine include researching efficient model inference at the System & Artificial Intelligence (SAI) Lab with Prof. Sai Zhang, working on multilingual representation learning at the National University of Singapore with Prof. Liu Lili, and leading robotics control work as part of the IIT Bombay E-Yantra Robotics initiative.

I also hold a Bachelor’s Degree in Computer Science Engineering, where I explored reinforcement learning and NLP.

Email  /  CV  /  Google Scholar  /  Github  /  Music /  Twitter

Research

I am broadly interested in multimodal representation learning, world modeling, and perception, with the motivation to build agents in robotics and healthcare that can understand, adapt, and generalize across diverse environments.

Introduces a GAN-augmented ensemble model using plasma cfRNA to predict CSF pressure non-invasively. Demonstrates high predictive accuracy and clinical relevance for ICP monitoring in spaceflight and clinical contexts.

Comparative study of GAN variants (DCGAN, ProGAN, StyleGAN2) for anime face synthesis. StyleGAN2 demonstrated superior effectiveness, achieving the lowest FID score among tested models.

Review highlighting the role of AI-enhanced diffusion MRI and PET imaging in early diagnosis of neurodegenerative and neuro-ophthalmic disorders, integrating multimodal data with deep learning models for improved precision.

Teaching

Template adapted from Jon Barron's website