Use LLMs to judge quality of other LLM output

Use LLMs to judge quality of other LLM output

Cameron Wolfe Using LLMs for Evaluation (archive link)