A passing evalA repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.Full definition → suite is not evidence the suite would catch a regression, and mutation scoring is a concrete way to measure that gap.
Terms in this piece · Glossary
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
articlesmevals - a small eval suite for evaluating models, prompts, and harnessesSimon Willison
articleBreaking Models to Test the Judge: A Mutation Testing Approach for Semantic Evaluators of Domain Class DiagramsKevin Delcourt, Meriem Ben Chaaben, Abdelhamid Rouatbi, Luciano Marchezan, Houari Sahraoui
articleWhen Knowledge Changes: Metamorphic Testing of RAG Systems with MutationsJinhan Kim, Samuele Pasini, Paolo Tonella