🍊 Latent Atlas 🍉

标签: instructgpt

此标签下有1条笔记。

  • 2026年5月29日

    Training language models to follow instructions with human feedback

    • source
    • paper
    • instructgpt
    • rlhf
    • sft
    • reward-model

🍊 Latent Atlas 🍉 · An AI knowledge atlas built with Quartz © 2026