Frontier Safety Eval Loop

提示词

Create a beautiful research flowchart for an AI safety benchmark pipeline called Frontier Safety Eval Loop. Landscape figure, white background, large typography, vector-like shapes, soft indigo, coral, sage, and graphite palette. Show stages Prompt Suite, Model Runs, Judge Models, Human Audit, Failure Taxonomy, Patch Queue, and Re-run. Use clean swimlanes, numbered callouts, compact legends, and premium paper-ready styling. High detail, excellent color harmony, generous whitespace, no clutter, conference-quality diagram.

中文译文

为一条名为 Frontier Safety Eval Loop 的 AI 安全基准测试管线创作一张精美的研究流程图。横向画幅，白色背景，大号排版，矢量感图形，柔和靛蓝、珊瑚、鼠尾草绿与石墨色调色板。展示 Prompt Suite、Model Runs、Judge Models、Human Audit、Failure Taxonomy、Patch Queue 和 Re-run 各阶段。使用干净的泳道、编号标注、紧凑图例以及高级的论文级排版风格。高细节，色彩和谐出色，留白充裕，无杂乱，会议级图表。

溯源

原作者

wuyoscar

收录来源

wuyoscar/GPT-Image2-Skill · 92 · MIT