Shalini Ghosh

A new framework automates testing generative models for unsafe content.

A new method enhances speech recognition efficiency using low-rank adaptation.

LLMs enhance accuracy and error correction in speech recognition systems.