发表机构
Northeastern University; West Virginia University（东北大学; 西弗吉尼亚大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究基于预测驱动推理提出预测驱动评估框架，引入预测驱动节省率元指标，结合有限人工判断与大规模自动评分实现无偏高效的系统比较，可节省人工标注成本，适用于各类无法验证的任务。
AI 中文摘要
在各类无法验证的任务中，人工评估可靠但成本高昂，而自动指标更具可扩展性但往往存在偏差。基于预测驱动推理（Prediction-Powered Inference, PPI），我们提出预测驱动评估框架，该框架将有限的人工判断与大规模自动评分相结合，以实现数据高效且可证明无偏的系统比较。我们开发了参数化与非参数化方法，分析配对与非配对设计之间的效率权衡，并在六个WMT数据集上验证了该框架。我们进一步引入预测驱动节省率（Prediction-Powered Saving Ratio, PPSR），这是一种元指标，用于衡量自动指标在预测驱动评估中可节省的人工标注量。PPSR直接针对预测驱动评估的指标效用，比现有的系统级元指标产生更具区分度和稳定性的指标排名。总体而言，我们的新范式将自动指标重新定义为降低人工标注成本的工具，而非替代人工判断，且广泛适用于各类无法验证的任务。
英文摘要
Across various non-verifiable tasks, human evaluation is reliable but expensive, while automatic metrics are more scalable but often biased. Building on prediction-powered inference (PPI), we propose prediction-powered evaluation, a framework that combines limited human judgments with large-scale automatic scores to obtain data-efficient system comparisons that are provably unbiased. We develop parametric and non-parametric procedures, analyze the efficiency trade-off between paired and unpaired designs, and validate the framework on six WMT datasets. We further introduce the Prediction-Powered Saving Ratio (PPSR), a meta-metric that measures how much human annotation an automatic metric can save when used within prediction-powered evaluation. PPSR directly targets metric utility for prediction-powered evaluation and yields more discriminative and stable metric rankings than existing system-level meta-metrics. Overall, our new paradigm reframes automatic metrics as tools for reducing human annotation cost rather than replacing human judgment, and applies broadly to non-verifiable tasks.
CommentsAccepted at EMNLP 2026 (Main). Code available: https://github.com/CHATS-lab/ppi-eval
