论文进阶
Rubric
评分量规，结构化的打分指南，定义评估维度、评判标准与对应质量等级，用于系统化打分。
English Definition
"A structured scoring guide that defines evaluation dimensions, criteria, and quality levels used to rate outputs systematically."
用法说明 · 针对中文母语者
在 HCI/UX 评估中常与启发式评估、可用性测试搭配使用；中文常译作 '评分量规' 或 '评分准则'。
真实用例 · 2 条
"We conduct a fine-grained evaluation of three LLMs using both human annotators and LLM-as-judge methods, utilizing rubrics created in consultation with CSOs."
论文Pipeline extracted — review needed·说明人工与机器评审共用同一份与 CSO 共同制定的量规
"Each model is evaluated along multiple dimensions: clarity & fluency, helpfulness & relevance, completeness & conciseness, and accuracy."
论文Pipeline extracted — review needed·列出量规所覆盖的具体评估维度
由 pipeline 自动采集，待人工 review