News for hackers. Stuff that matters.

A comment on Reasoning models don't always say what they think. Parent comment

Re:Reasoning models don't always say what they think by dragonwriter on Thursday April 03, 2025 @02:30PM (#43573557)
No, what was described was specifically reporting to an external party the neural connections involved in the mistake and the source in past training data that caused them, as well as learning from new data.

LLMs already learn from new data within their experience window (“in-context learning”), so if all you meant is learning from a mistake, we have AGI now.

[ Parent | on HN ]
Re:Reasoning models don't always say what they think by Jensson on Thursday April 03, 2025 @05:47PM (#43575865)
> LLMs already learn from new data within their experience window (“in-context learning”), so if all you meant is learning from a mistake, we have AGI now.

They don't learn from the mistake though, they mostly just repeat it.

[ Parent | on HN ]