Assistant Professor, McGill University
1 paper at NeurIPS 2025
In this position paper we investigate the validity and reliability of LLMs as judges and highlight challenges inherent to their use and existing practices in NLG evaluation.
