• Feathercrown@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    16 days ago

    This unironically works, it’s basically the same reason why chain-of-reasoning models produce better outputs