Unpacking the HF in RLHF

Re:Unpacking the HF in RLHF by jbcranshaw on Thursday March 09, 2023 @04:10PM (#35086762)

Some observations on a few ways different people actually gather feedback from humans in practice to improve LLMs. Sure I've missed some here, so let me know.

[ Parent | on HN ]

Re:Unpacking the HF in RLHF by PaulHoule on Thursday March 09, 2023 @04:45PM (#35087174)

I think tying the feedback to a task is the way to go. I don't know exactly what they do with it but my son is a fan of Craiyon and Craiyon shows you a few images it generated and encourages you to favorite the best ones. I'm sure you could RLHF an image generator too.

See my comment on this thread

https://news.ycombinator.com/item?id=35069965

I'd be glad to chat more (see my profile) but I think as much as people think there is a scalability advantage to a big co training one big model the problems of pleasing everybody, particularly advertisers, are terrible but a personal model that pleases one person might be easy with recent tech.

[ Parent | on HN ]

Re:Unpacking the HF in RLHF by jbcranshaw on Thursday March 09, 2023 @04:47PM (#35087205)

Hey, thanks for the reply. Would love to chat more.

[ Parent | on HN ]