Learned-Rule-Augmented Large Language Model Evaluators

Jie Meng, Jin Mao

Large language models (LLMs) are predominantly used as evaluators for natural language generation (NLG) tasks, but their application to broader evaluation scenarios remains limited…