返回探索

Jev 与六个 LLM 玩贪吃蛇对比实验

作者用相同种子和四向决策空间,让 Jev 与六个 LLM 各玩 60 秒贪吃蛇,Jev 获得 21 分并做出 204 次决策。

“Models have been superhuman at chat for years. So where is all the automation?” I took @CompleteSkeptic’s question literally and built a small experiment. Jev vs. six LLMs playing Snake. Same seed, same four-way decision space, 60 seconds each. Jev: 21 pts / 204 decisions

· 0 次赞在 X 打开