FourWaveforms @ FourWaveforms @lemm.ee

I don't mean to be difficult. I'm neurodivergent

---

Read more

Posts

0

Comments

317

Joined

4 mo. ago

If you don't play chess, the Atari is probably going to beat you as well.

LLMs are only good at things to the extent that they have been well-trained in the relevant areas. Not just learning to predict text string sequences, but reinforcement learning after that, where a human or some other agent says "this answer is better than that one" enough times in enough of the right contexts. It mimics the way humans learn, which is through repeated and diverse exposure.

If they set up a system to train it against some chess program, or (much simpler) simply gave it a tool call, it would do much better. Tool calling already exists and would be by far the easiest way.

It could also be instructed to write a chess solver program and then run it, at which point it would be on par with the Atari, but it wouldn't compete well with a serious chess solver.

That's just impolite.

maybe we should ban all the vices you don't partake of but leave the rest intact

depends on which marketer you ask

ive been staring at this for 20 minutes. when does it end

...in Singapore...

My wife left me... and so did my knives, it turns out

I've seen models do this in benchmarks. It's how they respond without reinforcement learning. Also this is probably fake

I think it would be best if that's a user setting, like dark mode. It would obviously be a popular setting to adjust. If they don't do that, there will doubtless be grease monkey and other scripts to hide it.

I wasn't trolling

I like it

I can strapolate all the stats are what he made up in his head

Many people who are in a position to write opaque jargon lack the perspective that would be required to explain it to a person who isn't already very well-versed. Math articles are often like that, which doesn't surprise me. I've had a few math professors who appeared completely unable to understand how to explain the subject to anyone who wasn't already good at it. I had to drop their classes and try my luck with others.

I feel like a few of them are in this thread!

Why do you think their reading skills are poor?

I don't trust even the best modern commercial models to do this right, but with human oversight it could be valuable.

You're right about it being a lost fight, in some ways at least. There are lawsuits in flight that could undermine it. How far that will go remains to be seen. Pissing and moaning about it won't accelerate the progress of those lawsuits, and is mainly an empty recreational activity.

Then skip the AI summary.

Which Dijkstra quote?

https://youtu.be/LB5Wlv_AHpg

It could be Sanka, which is p. much the same

this guy has everything figured out