There is an increasing trend towards evaluating NLP models with LLM-generated
judgments instead of human judgments. In the absence of a comparison against
human data, this raises concerns about the validity of these evaluations; in
case they are conducted with proprietary models, this also raises concerns over
reproducibility. We provide JUDGE-BENCH, a collection of 20 NLP datasets with
human annotations, and comprehensively evaluate 11 current LLMs, covering both
open-weight and proprietary models, for their ability to replicate the
annotations. Our evaluations show that each LLM exhibits a large variance
across datasets in its correlation to human judgments. We conclude that LLMs
are not yet ready to systematically replace human judges in NLP.

评估 NLP 模型时，使用 LLM-generated 判断取而代之人为判断的趋势日益增长。我们提供了一个由人工注释的 20 个 NLP 数据集的 JUDGE-BENCH，并对 11 个当前的 LLM 进行全面评估，涵盖公开权重和专有模型，以验证其模拟注释的能力。我们的评估结果表明，每个 LLM 在与人工判断的相关性方面在不同数据集上存在很大的差异。我们得出结论，LLM 尚不具备系统替代 NLP 中的人类评审员的能力。