swyx@swyx · XAI 摘要
可能是我见过的最好的推理效率奖励函数。
probably the best reward function for reasoning efficiency i've seen
length penalty is very elegant and simple tbh
swyx@swyx46
AI 摘要
可能是我见过的最好的推理效率奖励函数。
probably the best reward function for reasoning efficiency i've seen
length penalty is very elegant and simple tbh
swyx@swyx · X在 X 查看原推x.com