It's Not Just X. It's Y

(mail.cyberneticforests.com)

33 points | by mooreds 1 hour ago

5 comments

  • Baader-Meinhof 8 minutes ago
    I like that these AI idioms exist. They're like watermarks for text. It's worth the cost of humans avoiding them. Companies will eventually train their models to be undetectable, but society would be better if they didn't.
  • wrs 6 minutes ago
    This is how early forms of "reasoning" in LLMs worked: just literally inserting words like "Wait...", "Hmm...", "Let me reconsider...", "But is it really..." into the token stream.
  • Retr0id 18 minutes ago
    > RLVR is weirder, and I suspect it's why we see "It's not X, it's Y" so often.

    This feels like an easy enough hypothesis to verify, for anyone in the business of training LLMs - does the not-X-but-Y rate increase after RLVR?

    • andy99 2 minutes ago
      It’s unlikely this is true. LLMs are way more mad-libs / templates than we like to admit, that’s (ironically) not a judgement about their capability, it’s primarily just an observation. But it’s also what plain old SFT, which I believe is the primary culprit, ends up imparting.
  • rvz 12 minutes ago
    Another bunch of dead give aways in code bases with READMEs is the repetitive:

    - "No X, No Y, No Z." pattern

    - "Here is X - it makes Y"

    The worst and most obvious one is the constant over use of emoji ticks and crosses.

  • huflungdung 16 minutes ago
    You’re absolutely right. This is the smoking gun. This changes everything.
    • Starlevel004 2 minutes ago
      This is the real unlock. Here's the key takeaways.