Berkeley AI Research (BAIR)

My first time in academia. Published thesis centered around analyzing weakness in embedding models in unclean data environments, and improving text similarity measurements to be robust and aligned with human understanding.

Currenty working on 2 Neurips submissions focused on embedding models and text similarity measurement in LLMs.

Masters Thesis: https://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-84.html