New comment by qwery2 in "GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2"

qwery2 — Sun, 21 Jun 2026 01:06:45 +0000

RLVR is a process which updates the Markov chain

New comment by qwery2 in "What can LLMs never do?"

qwery2 — Sun, 28 Apr 2024 01:34:04 +0000

I think the person you're replying to may have been referring to the problem of a MLP approximating a sine wave for out of distribution samples, i.e. the entire set of real numbers.

Hacker News: qwery2

New comment by qwery2 in "GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2"

New comment by qwery2 in "What can LLMs never do?"