MultEval: Supporting Collaborative Alignment for LLM-as-a-Judge Evaluation Criteria
For researchers and practitioners using LLM-as-a-judge evaluation, this work addresses the overlooked collaborative and contested nature of criteria definition, but the contribution is incremental as it focuses on tool support rather than a new evaluation paradigm.
The paper identifies challenges in collaborative creation of evaluation criteria for LLM-as-a-judge systems, including difficulties in establishing shared understanding and aligning values. It introduces MultEval, a system that supports collaborative criteria authoring through consensus-building and transparency, demonstrated via a case study with domain experts.
LLM-as-a-judge approaches have emerged as a scalable solution for evaluating model behaviors, yet they rely on evaluation criteria often created by a single individual, embedding that person's assumptions, priorities, and interpretive lens. In practice, defining such criteria is a collaborative and contested process involving multiple stakeholders with different values, interpretations, and priorities; an aspect largely unsupported by existing tools. To examine this problem in depth, we present a formative study examining how stakeholders collaboratively create, negotiate, and refine evaluation criteria for LLM-as-a-judge systems. Our findings reveal challenges in human oversight, including difficulties in establishing shared understanding, aligning values across stakeholders with different expertise and priorities, and translating nuanced human judgments into criteria that are interpretable and actionable for LLM judges. Based on these insights, we developed MultEval, a system that supports collaborative criteria by enabling multiple evaluators to surface and diagnose disagreements using consensus-building theory, iteratively revise criteria with attached examples and proposal history, and maintain transparency over how judgments are encoded into an automated evaluator. We further report a case study in which a team of domain experts used MultEval to collaboratively author criteria, illustrating how coordination and collaborative consensus-making shape criteria evolution.