Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Karpathy’s Pelican

twitter.com

Post reply on HN
Print view View original

101–110 of 461 posts

Re: Karpathy’s Pelican

#101

Post by egeozcan »

tripleee wrote: post #78
8n4vidtmkvmk wrote: post #57

Earlier quoted context omitted.

We're all just impressed that it's even possible. No one is saying these are fun. I guess the next question is if they can be made fun without too much additional work with a human guiding the AI.

> the next question is if they can be made fun without too much additional work They can't. If you think about how these things are trained it's blatantly obvious fun is an impossible metric to optimize them for

Training impossibilities aside, how would you even make an optimization loop for "fun"?

Re: Karpathy’s Pelican

#103

Post by dekhn »

I'd like to see the Silmarillion, specfically both Ainulindalë and the Fall of Numenor. At this point a visual model would probably produce something better than Amazon (but presumably not Jackson).

Re: Karpathy’s Pelican

#104

Post by croes »

> I also like this kind of examples because no one in their right mind would ever spend the time to write something this custom

There are people in their right mind who would do that and their are already examples of people who did similar things.

But maybe not in the future if people would confuse all the effort with AI

Re: Karpathy’s Pelican

#105

Post by upmostly »

matsemann wrote: post #41

I'm pretty tired of the "Y made this game in Z tokens" all over the internet last week. They look impressive, and it's cool that it's even possible, but they're useless as games. None of them are any fun. They're like the most boring variant of basic controllers you can imagine. None have any cool mechanics. None have any tweaks made from hours and hours of testing. All have the same cel-shader.

Unfortunately the Phaser framework has decided to go all-in on this route, and it's really disappointing.

These one-shot products aren't games. They're barely even demos. I don't even know what to call them. For a mature framework like Phaser to sell-out like this and create a vibecoded platform for vibecoded games is shocking.

Re: Karpathy’s Pelican

#106

Post by dofm »

fwlr wrote: post #99

I really dislike this AI programming thing of “Mr LLM, go slam your face into the problem until there’s no problem left, then call me back”. (Not sure if it’s a recent trend or a fundamental nature.) It always brings to my mind some words from Rich Hickey: I think we’re in this world I’d like to call “guardrail programming”. It’s really sad: we’re like, “I can make change because I have tests!”. Who does that? Who dr…

> I really dislike this AI programming thing of “Mr LLM, go slam your face into the problem until there’s no problem left, then call me back”.

Not difficult to see why the employees of AI firms are thrilled with it though, eh?

Re: Karpathy’s Pelican

#107

Post by mdp2021 »

rzzzt wrote: post #102
mdp2021 wrote: post #68

Earlier quoted context omitted.

Can someone please translate that expression? What would that mean?

Affirmation?

Well, in that case - if it is just a "hear, hear" - I do not see why the utterance from that actor would be of note.

Maybe nozzlegear wanted to suggest some importance on Musk remaining a bet-ter on the general tech, regardless of the competition?

Re: Karpathy’s Pelican

#108

Post by dofm »

Anthropic spokesman [0] Andrej Karpathy is here to tell you about token-wasting loops, and insists on the weird idea that they are "~free", when in fact, they are fuelled by expensively burning investor money.

[0] Seriously. Get used to mentally prefixing his and Boris Cherny's name like this, every time you see them quoted. These people are speaking while employed; there is no chance they are not aligned with the employers who will make them wealthy. The tech industry does like to pretend that for some reason AI people, uniquely, speak thoughts unbiased and for themselves or even for science or humanity.

Re: Karpathy’s Pelican

#109

Post by maxutility »

jmugan wrote: post #93

A lot of people are posting here about how bad the end product is, but that is kind of the point. Models have moved beyond generating images to a new kind of benchmark that better exposes understanding of the physical world, and we can use benchmarks like this to measure future progress. (Of course, it will have to be a qualitative/subjective measurement.)

Agree. The pelican benchmark was interesting a year ago when most models struggled and a good pelican indicated an unusually capable model. Now it’s saturated and uninteresting.

A good new benchmark should have awful performance to start and there should be a lot of headroom for improvement. This benchmark is also intentionally difficult and requires the LLM to develop the animation through spatial reasoning and first principals rather than existing video generation pipelines. Similar to how SVG generation was out of distribution for most models a year ago.

Re: Karpathy’s Pelican

#110

Post by tripleee »

egeozcan wrote: post #101
tripleee wrote: post #78

Earlier quoted context omitted.

> the next question is if they can be made fun without too much additional work They can't. If you think about how these things are trained it's blatantly obvious fun is an impossible metric to optimize them for

Training impossibilities aside, how would you even make an optimization loop for "fun"?

Brain-computer interface, maybe. Hook a million play-testers up to the output and have it iterate.

..feels like there was a black mirror episode about that though

Post reply on HN