Lmunit: Fine-grained evaluation with natural language unit tests

[Free GPT-4]
[PDF] arxiv.org

Learning to Plan & Reason for Evaluation with Thinking-LLM-as-a-Judge

S Saha, X Li, M Ghazvininejad, J Weston… - arxiv preprint arxiv …, 2025 - arxiv.org
LLM-as-a-Judge models generate chain-of-thought (CoT) sequences intended to capture
the step-bystep reasoning process that underlies the final evaluation of a response …
Save Cite Related articles View as HTML