Rahul Nair

A new method to refine reward systems in reinforcement learning using user input.

Introducing a novel approach to rank language models without needing reference answers.

Addressing the challenges of geographic bias in object recognition technology.

Bias in datasets can worsen AI predictions, leading to unfair outcomes.