Evaluator-Optimizer: Iterative Refinement with a Separate Critic LLM
·7 min·AI
Some tasks improve on a second pass, but only when the critic is a separate LLM with its own mandate. Asking a generator to grade its own work produces confident praise for mediocre output.