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

201–210 of 461 posts

Re: Karpathy’s Pelican

#201

Post by twostorytower »

YmiYugy wrote: post #172

I don't think it's a bad way to benchmark new models, I just find it concerning that the author implies that "pelican on a bicycle" has been exhausted. At the risk of making overly broad, unfalsifiable claims I think multi-year exposure to AI content has dramatically raised our expectations for speed and volume but lowered them for quality. We see a very janky pelican and declare the problem solved.

100%. Why waste the tokens to render Lord of the Rings when the pelican test still clearly benchmarks so well.

Re: Karpathy’s Pelican

#205

Post by NewJazz »

jcims wrote: post #202

I’d like to see a human one shot a pelican on a bicycle in raw svg.

What do you mean by "oneshot"? The term applied to genai makes sense, but it doesn't make a whole lot of sense to apply the term to human art.

Re: Karpathy’s Pelican

#206

Post by matsemann »

segmondy wrote: post #146
throwatdem12311 wrote: post #94

Earlier quoted context omitted.

It’s the people spamming these shitty games that lack imagination.

people are excited, let them be excited! but more than the excitement is realizing the implications of what this means for building software, first it looks like a toy and then it doesn't. the OP is talking about how the games are not playable and fun. who cares? that's not the point.

What does it mean for building software? These games cannot become fun. There is no way to prompt it fun, and any change you try to prompt it to make will inevitably blow up something else. The code is a mess and impossible to build upon.

Re: Karpathy’s Pelican

#207

Post by anigbrowl »

dundarious wrote: post #51

I can forgive the modeling being godawful jank (windows floating in the air, disconnected from the house). But I expected it to have a better understanding of the text. Instead, we have Bilbo's "disappearance" interpreted as him magically transporting or cloaking, and similarly for his reappearance.

But this only seems wrong to you because you're familiar with the prior context. When there's only a single paragraph to work from, and it's a drily humorous text, why not employ comic literalism and lean into the perplexity with which his neighbors viewed him?

Re: Karpathy’s Pelican

#209

Post by jonas21 »

YmiYugy wrote: post #172

I don't think it's a bad way to benchmark new models, I just find it concerning that the author implies that "pelican on a bicycle" has been exhausted. At the risk of making overly broad, unfalsifiable claims I think multi-year exposure to AI content has dramatically raised our expectations for speed and volume but lowered them for quality. We see a very janky pelican and declare the problem solved.

I don't think he's claiming it's been exhausted. It's just that things have progressed to a point where people are arguing over the finer points of which pelican looks better -- which is often a matter of taste, and an indication that we've hit the knee in benchmark where models are no longer failing in obviously awful ways.

Re: Karpathy’s Pelican

#210

Post by djhworld »

It would be interesting to see the models work on a book it hasn't been trained on yet. I guess sadly that means any book released very recently.

Definitely impressive demo, I do wonder though if the countless artwork, films, images etc produced over many decades around Lord of the Rings somewhat influenced the outcome of this though.

Post reply on HN