Back to Reading Room

LA BIBLIOTHÈQUE · READING NOTE

Training Language Models to Follow Instructions with Human Feedback

Ouyang et al.

To Read · 想看

AIAlignmentPapers

InstructGPT, RLHF, Alignment, Instruction Following

InstructGPT 论文,建立经典的 SFT、偏好数据、Reward Model 与 RLHF 流程,是理解现代 Chat Assistant 如何从基础模型变成指令跟随模型的关键论文。