@inproceedings{lee-etal-2025-checkeval,
    title = "{C}heck{E}val: A reliable {LLM}-as-a-Judge framework for evaluating text generation using checklists",
    author = "Lee, Yukyung  and
      Kim, JoongHoon  and
      Kim, Jaehee  and
      Cho, Hyowon  and
      Kang, Jaewook  and
      Kang, Pilsung  and
      Kim, Najoung",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.796/",
    pages = "15782--15809",
    ISBN = "979-8-89176-332-6",
    abstract = "Existing LLM-as-a-Judge approaches for evaluating text generation suffer from rating inconsistencies, with low agreement and high rating variance across different evaluator models. We attribute this to subjective evaluation criteria combined with Likert scale scoring in existing protocols. To address this issue, we introduce CheckEval, a checklist-based evaluation framework that improves rating reliability via decomposed binary questions. Through experiments with 12 evaluator models across multiple datasets, we first demonstrate that CheckEval strongly correlates with human judgments. More importantly, CheckEval dramatically improves the average agreement across evaluator models by 0.45 and reduces the score variance. CheckEval scores furthermore have the benefit of being more interpretable because it decomposes evaluation criteria into traceable binary decisions, allowing analyses of specific attributes driving quality judgments."
}