News for hackers. Stuff that matters.

A comment on Learning to Reason with LLMs.

Re:Learning to Reason with LLMs by joshhug on Thursday September 12, 2024 @06:56PM (#41526499)
I just tried o1, and it did pretty well with understanding this minor issue with subtitles on a Dutch TV show we were watching.

I asked it "I was watching a show and in the subtitles an umlaut u was rendered as 1/4, i.e. a single character that said 1/4. Why would this happen?"

and it gave a pretty thorough explanation of exactly which encoding issue was to blame.

https://chatgpt.com/share/66e37145-72bc-800a-be7b-f7c76471a1...

[ Parent | on HN ]
Re:Learning to Reason with LLMs by piker on Thursday September 12, 2024 @07:01PM (#41526547)
A common problem, no doubt, with a lot of training context. But man. What a time to be alive.
[ Parent | on HN ]
Re:Learning to Reason with LLMs by takoid on Thursday September 12, 2024 @07:07PM (#41526589)
4o’s answer seems sufficient, though it provides less detail than o1.

https://chatgpt.com/share/66e373d7-7814-8009-86c3-1ce549ca2e...

[ Parent | on HN ]
Re:Learning to Reason with LLMs by karmasimida on Thursday September 12, 2024 @07:07PM (#41526591)
Damn, the model really goes to length to those trivial but hard problems. Impressive
[ Parent | on HN ]