TechTakes

1401 readers

204 users here now

Big brain tech dude got yet another clueless take over at HackerNews etc? Here's the place to vent. Orange site, VC foolishness, all welcome.

This is not debate club. Unless it’s amusing debate.

For actually-good tech, you want our NotAwfulTech community

founded 1 year ago

MODERATORS

[email protected]

200

LLMs can’t reason — they just crib reasoning-like steps from their training data (pivot-to-ai.com)

submitted 1 month ago by [email protected] to c/[email protected]

129 comments fedilink hide all child comments

you are viewing a single comment's thread
view the rest of the comments

[–] [email protected] 13 points 1 month ago (3 children)

Isn’t OpenAI saying that o1 has reasoning as a specific selling point?

[–] [email protected] 14 points 1 month ago (1 children)

they do say that, yes. it’s as bullshit as all the other claims they’ve been making

[–] [email protected] 8 points 1 month ago

Which is my point, and forgive me, but I believe is the point of the research publication.

[–] [email protected] 12 points 1 month ago

They say a lot of stuff.

[–] [email protected] 6 points 1 month ago (1 children)

My best guess is it generates several possible replies and then does some sort of token match to determine which one may potentially be the most accurate. Not sure if I'd call that "reasoning" but I guess it could potentially improve results in some cases. With OpenAI not being so open it is hard to tell though. They've been overpromising a lot already so it may as well be just complete bullshit.

[–] [email protected] 4 points 1 month ago (1 children)

My best guess is it generates several possible replies and then does some sort of token match to determine which one may potentially be the most accurate.

Didn't the previous models already do this?

[–] [email protected] 4 points 1 month ago

No idea. I'm not actually using any OpenAI products.