design-ai-benchmarking
@aperivue · 收录于 1 周前
Design and validity review for studies that benchmark one or more AI systems against a human-expert panel as the reference. Covers the evaluation question and arm definition, decoupled multi-dimensional rubrics with anchors, planted calibration probes, reviewer-panel construction, inter-rater reliability targets, LLM-as-judge versus human-as-judge adjudication, construct-independence guards, and a structured rating-export schema. Use before data collection on an AI-vs-expert evaluation.
适合你，如果你需要设计AI系统与人类专家对比的评估研究
/ 通过 npx 安装 校验哈希
npx oh-my-skill add aperivue/medsci-skills/design-ai-benchmarking/ 通过 bash 安装
curl -fsSL https://oh-my-skill.com/install.sh | bash -s -- aperivue/medsci-skills/design-ai-benchmarking/ 已经装过?验证本机副本,不用重装
npx oh-my-skill verify aperivue/medsci-skills/design-ai-benchmarking安装目标可用 --agent / --scope 或 --to 明确指定;省略时只会在唯一已存在的 agent 目录上自动选择,零命中或多命中会停止并提示。content_hash 缺失或不一致均拒装。
190GitHub stars
~2.4K最小装载
~2.4K含声明引用
~4.7K文本包总量
镜像托管
评论
登录即可评论;带「已验证安装」的,是发布者名下有本店的安装或持有记录。
…