MattRix

26 ms·

421.
by MattRix 3y ago
It’s not that hard to get these LLMs to generate new game rules, I’ve made some prototypes that use them to do that. The hard part is getting them to generate rules that are actually fun to play and somewhat balanced. The fact is that it’s
422.
by MattRix 3y ago
There are lots of models/approaches to look at if you want improved speed. It seems bizarre to not also want the speed to increase.
423.
by MattRix 3y ago
You’re reading something into the text that just isn’t there. “Rival” means it’s good enough to be competitive, it doesn’t mean it’s better. The statement is hardly surprising or controversial. It’s fairly obvious to anyone who has looked a
424.
by MattRix 3y ago
The first graph on the page compares the quality of it, showing that it has higher quality results despite taking only 2 seconds vs results from Stable Diffusion that take 1.4 minutes.
425.
by MattRix 3y ago
I don’t think you could ever accurately simulate trash movement. Trash is irregularly shaped, even with the highest resolution cameras, it’d be impossible to know what the far side of the trash is shaped like. Then on top of that there’s no
426.
by MattRix 3y ago
Yeah that’s nowhere close. It’s easy to make a prototype that looks good in a marketing video while driving a very tightly mapped route. It’s a whole other thing to let anyone use self driving tech anywhere, especially on routes it has neve
427.
by MattRix 3y ago
Some of those stories are tragic, but they are a tiny fraction of the number of people who would literally given up anything they own (including their own life) for their kids.
428.
by MattRix 3y ago
“ Is he even serious about any of this?” Not sure how people are misunderstanding him, he literally said he was joking… He knows it’s not a great deal for Reddit. His whole point is that the app isn’t actually worth $20 million a year, whic
429.
by MattRix 3y ago
A million over five years.
430.
by MattRix 3y ago
Yes, that’s why I am saying they will have to ensure the LLM doesn’t do that.
431.
by MattRix 3y ago
The reciting part is illegal, but as long as it is trained not to recite things in full (or to whatever limit the law determines), then it should be fine.
432.
by MattRix 3y ago
They aren’t saying that AI needs to stop existing, just that more effort needs to be put towards regulating it, monitoring it, aligning it, etc.
433.
by MattRix 3y ago
There are still lots of good tech and design talks, it’s just that as you get older and more experienced, your standards for what is interesting and worth your time get more strict.
434.
by MattRix 3y ago
I don’t see how it’s doublespeak, they only started calling it that after opening it up for everyone to use. And it IS the most common charging standard in North America by far, both in terms of number of cars and number of chargers.
435.
by MattRix 3y ago
I think it should be censored because A. I think it’s important to protect victims of hate. and B. I believe allowing hate speech leads to hateful groups recruiting more members and gradually gaining power. Who decides what hate speech is?
436.
by MattRix 3y ago
The first Push was released in 2013, the Linnstrument in 2014.
437.
by MattRix 3y ago
As long as cyclists are forced to share the road with cars they should get a say when it comes to car safety, especially when it comes to properties of cars that make them more dangerous to cyclists and others on the road. Also worth pointi
438.
by MattRix 3y ago
Are hate speech censorship apologists really the worst? I consider myself someone who believes hate speech should be censored. I’m pretty sure there are much worse things.
439.
by MattRix 3y ago
Apple’s problem has never been the “barrier to entry”. They know developers will make apps for their platform no matter what. If Swift becomes more universal it makes it easier for Apple developers to also develop for Android (etc) which Ap
440.
by MattRix 3y ago
It’s really not that hard to make good looking stuff with these AI image generators, you just need the right models. The reason you struggled is because the basic stable diffusion model creates really ugly stuff by default. Try using a mode
441.
by MattRix 3y ago
I see what you’re describing as investing, whereas the speculation is the gambling part of it.
442.
by MattRix 3y ago
I mean speculation IS just another type of gambling, yes.
443.
by MattRix 3y ago
Rooms seems like a very different style of product (not built around the social/chatting stuff). Also those kind of isometric hexagon rooms have been around for ever in loads of different places.
444.
by MattRix 3y ago
And?
445.
by MattRix 3y ago
Why? It seems like something more CS students could learn about.
446.
by MattRix 3y ago
The vast majority of tweets about ChatGPT are the same garbage. I think it’s mostly due to Twitter giving too much weight to tweets and likes from “verified” users. I wouldn’t be surprised that if you looked at the details on those tweets,
447.
by MattRix 3y ago
Not sure it’s worth responding to this other than saying it’s disgusting to even pretend this is equivalent to slavery.
448.
by MattRix 3y ago
Yes it’s not predictable but not necessarily non-deterministic.
449.
by MattRix 3y ago
It’s not really an indirect charge, he did try to cover it up.
450.
by MattRix 3y ago
I think you’d be surprised at the number of photos posted, but also many people post tons of gifs (especially reaction gifs) which are fairly large.

More ›