Concept
AI-as-Judge
AI-as-Judge (also called LLM-as-Judge) is the practice of using a language or multimodal model to score, rank, or critique the output of another model, replacing or supplementing human evaluators. Think of it like unit tests for generated text, except the assertion is itself…
The rest of “AI-as-Judge” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→