Google Research·· 2026-04-01AI 评分47
Google Research 研究:构建更好的 AI 评测基准需要多少标注者?
Building better AI benchmarks: How many raters are enough?
AI 导读
Google Research 在论文 "Forest vs Tree: The (N,K) Trade-off in Reproducible ML Evaluation" 中研究了评测基准中条目数量与每条标注者人数的权衡。
来源:Google Research · research.google