Depending on the specificity of the research, having a model with fewer parameters will come with a higher penalty. If you want a model to perform better at something specific while staying smaller, generally it will take specific training to achieve that.
how much of a clean room can you claim when you don't know exactly what code has your LLM looked at?
love the project of course, but LLMs are a huge caveat to such claims, which will be very hard to make credibly in the future for anything not entirely novel
We'll get to the point, if we're not already there, where you won't be able to tell if the artist actually did the work or just could have done it, and to which extent. Everything in the process can be essentially faked. If you put a massive emphasis on proving human work, you're essentially conceding you cannot tell without some sort of notary certification. We're in the lab diamond stage and clutching at some artificial authenticity.
people really believe this coordinated push across jurisdictions is about kids and verifying their age? this excuse to try to end pseudonimity on the web is as old as the mainstream internet itself
to a lot of people it never sat well that people could just go online and say whatever they want, and communicate with each other unsupervised at large scale, and be effectively untargetable while doing so - that model of the internet was only allowed because it happened under the radar and those uncomfortable with it have been fighting it since they got the memo
it's very odd that someone with no experience would take a big project like this and just jump to another language because he trusts the AI generated code of current models
if it works it works i guess, but it seems mad to me on the surface
Why do you think the creator behind SerenityOS has no experience? I mean it’s not the most popular OS out there but he seems like a capable individual.
in case it's not glaringly obvious from the comment, he has plenty of cpp experience and little rust experience, and that's according to his own comments
the relevant bit here is that he's porting from a language in which he has plenty of experience into another one in which he doesn't, in a large project
that in itself sounds like putting a lot of faith in LLMs but maybe there are factors not mentioned here, which is why i said "on the surface"
Indeed, the hard part won't the port, but the maintenance of that which got ported. To be fair though, he's probably going to be able to use the same techniques for that.
the trust element to me is jumping into a port, not specific code although code you didn't write in a language you're not an expert with, will ALWAYS introduce an added risk of falling into pitfalls you can only avoid with experience, the more the merrier
you're banking in the LLMs quite strongly when you do that, which may just work but i would be very worried about myself if i were in his boots
they failed to account for the fact that very similar effects are happening around the world in places where no investment was made to bring laptops to schools
these kids have smartphones and tablets and they spend countless hours on them, it's not that hard to see the effect this has
reply