logoalt Hacker News

dragonwriteryesterday at 10:26 PM2 repliesview on HN

That's inaccurate in two ways:

(1) The behavior that is approximately what you describe is not "fundamental" (though it may not be something you can disable on some hosted providers), it is an option that is not fundamental (and with runtimes where you have full control can be either disabled or tuned in a large number of manners), and

(2) The actual behavior that is approximately what you describe already usually involves use of PRNG (with a user or harness supplied seed), not a true RNG; the change to do watermarking isn't going from RNG to PRNG, it involves adding an additional set of constraints on token generation on top of the existing ones, which inherently compromises quality.


Replies

reliablereasonyesterday at 10:48 PM

(1) LLMs collapse and start outputting garbage after a number of tokens if you do not sample and just pick the "best token" each time. This is a consequence of how they are trained.

show 1 reply
beeringtoday at 12:43 AM

> which inherently compromises quality.

I don’t see how this follows? Tokens are chosen randomly. If you choose tokens with a different RNG in the same distribution, you’re still getting equally good or bad tokens.

show 1 reply