WorldBench：评估LLM生成Three.js体素世界

1. 12:00
   arXiv · Software Engineering

   WorldBench：评测 LLM 生成 Three.js 体素世界

   WorldBench 是一个面向开放式 LLM 生成 Three.js 体素世界的新基准与评测器，从一个含十个生物群落、物理及昼夜与季节循环的漂浮体素岛屿提示词出发，judge 控制时钟、环绕视角并派出 navigator agent 拍摄各生物群落，同时读取代码交叉核验。

Current heat 9·Comparable peak 10（Oct 9）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.