Hacker News
new
past
comments
ask
show
jobs
points
by
MattSayar
14 hours ago
|
comments
by
ACCount37
14 hours ago
|
[-]
Yeah, that's my point. Humans are not reliable LLM evaluators. "Secret model nerfs" happen in "vibes" far more often than they do in any reality.
reply