Sean Hendryx

Research uncovers concerns about large language models' math reasoning skills.

A novel approach to improve language models using human feedback.

A new method for detecting hallucinations in language models using corrupted data.