LiveFR

LLM-as-a-judge

Definition: LLM-as-a-judge uses a language model to score or compare another model's responses, against defined quality criteria.

It is a fast, scalable way to evaluate outputs hard to measure automatically. It stays imperfect: the judge can inherit biases and must be calibrated, often by cross-checking with human judgment.

See also

← Full AI glossary · AI news