News

Joined Google DeepMind as a Research Scientist in New York, working on understanding representations and how to control frontier multimodal models.

Completed my PhD at Georgia Tech with Vince Calhoun and Sergey Plis. Dissertation: The Hidden Structure of Deep Models: Discovering and Exploiting Sparse Subnetworks.

Learning to Query History accepted at the ICLR TSALM workshop, on retrieval-augmented classification under nonstationarity.

Concept Spaces in the Residual Stream of Diffusion Transformers accepted at the CVPR HOW workshop, on steering generation through concept directions.

SSFL: Discovering Sparse Unified Subnetworks at Initialization for Efficient Federated Learning was published in TMLR.

Joined Google DeepMind in Kirkland as a research intern with the Model Alignment, Control, Responsibility and Safety team.

Exploring Sparse Adapters for Scalable Merging of Parameter Efficient Experts was published at COLM.

Efficient Reinforcement Learning by Discovering Neural Pathways was accepted at NeurIPS.

Joined Cohere's LLM Efficiency Research team, working on inference-time activation sparsity for large language models.

Started a PhD research internship at Dolby Laboratories' Advanced Technologies Group, on parameter-efficient fine-tuning via probabilistic layer selection.