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

A new Google model is nearly perfect on automated handwriting recognition

generativehistory.substack.com

Post reply on HN
Print view View original

131–140 of 328 posts

Re: A new Google model is nearly perfect on automated handwriting recognition

#131

Post by hinkley »

fragmede wrote: post #99
scubbo wrote: post #62

Earlier quoted context omitted.

...what?

hinkley wrote: > We got the results back. You are a horrible person. I’m serious, that’s what it says: “Horrible person.” > We weren’t even testing for that. joshstrange then wrote: > If you want to listen to the line from Portal 2 it's on this page (second line in the section linked): https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Portal_2) ... as if the fact that the words that hinkley wrote are from a popular…

So if two sentences that make no sense to you sandwich one that does, you should totally accept the middle one at face value.

K.

Re: A new Google model is nearly perfect on automated handwriting recognition

#132

Post by roywiggins »

My task today for LLMs was "can you tell if this MRI brain scan is facing the normal way", and the answer was: no, absolutely not. Opus 4.1 succeeds more than chance, but still not nearly often enough to be useful. They all cheerfully hallucinate the wrong answer, confidently explaining the anatomy they are looking for, but wrong. Maybe Gemini 3 will pull it off.

Now, Claude did vibe code a fairly accurate solution to this using more traditional techniques. This is very impressive on its own but I'd hoped to be able to just shovel the problem into the VLM and be done with it. It's kind of crazy that we have "AIs" that can't tell even roughly what the orientation of a brain scan is- something a five year old could probably learn to do- but can vibe code something using traditional computer vision techniques to do it.

I suppose it's not too surprising, a visually impaired programmer might find it impossible to do reliably themselves but would code up a solution, but still: it's weird!

Re: A new Google model is nearly perfect on automated handwriting recognition

#133

Post by imiric »

famouswaffles wrote: post #89
imiric wrote: post #64

Earlier quoted context omitted.

Here's a thought experiment: if modern machine learning systems existed in the early 20th century, would they have been able to produce an equivalent to the theory of relativity? How about advance our understanding of the universe? Teach us about flight dynamics and take us into space? Invent the Turing machine, Von Neumann architecture, transistors? If yes, why aren't we seeing glimpses of such genius today? If we'v…

>Here's a thought experiment: if modern machine learning systems existed in the early 20th century, would they have been able to produce an equivalent to the theory of relativity? How about advance our understanding of the universe? Teach us about flight dynamics and take us into space? Invent the Turing machine, Von Neumann architecture, transistors? Only a small percentage of humanity are/were capable of doing any…

> LLMs are great, but they're not (yet?) as capable as our best and brightest (and in many ways, lag behind the average human) in most respects, so why would you expect such genius now ?

I'm not expecting novel scientific theories today. What I am expecting are signs and hints of such genius. Something that points in the direction that all tech CEOs are claiming we're headed in. So far I haven't seen any of this yet.

And, I'm sorry, I don't buy the excuse that these tools are not "yet" as capable as the best and brightest humans. They contain the sum of human knowledge, far more than any individual human in history. Are they not intelligent, capable of thinking and reasoning? Are we not at the verge of superintelligence[1]?

> we have recently built systems that are smarter than people in many ways, and are able to significantly amplify the output of people using them.

If all this is true, surely we should be seeing incredible results produced by this technology. If not by itself, then surely by "amplifying" the work of the best and brightest humans.

And yet... All we have to show for it are some very good applications of pattern matching and statistics, a bunch of gamed and misleading benchmarks and leaderboards, a whole lot of tech demos, solutions in search of a problem, and the very real problem of flooding us with even more spam, scams, disinformation, and devaluing human work with low-effort garbage.

[1]: https://blog.samaltman.com/the-gentle-singularity

Re: A new Google model is nearly perfect on automated handwriting recognition

#134

Post by mattmaroon »

mmaunder wrote: post #97

Substack: When you have nothing to say and all day to say it.

“This AI did something amazing but first I’m going to put in 72 paragraphs of details only I care about.”

I was thinking as I skimmed this it needs a “jump to recipe” button.

Re: A new Google model is nearly perfect on automated handwriting recognition

#135

Post by kazinator »

jstummbillig wrote: post #40
snickerbockers wrote: post #4

Earlier quoted context omitted.

I'm skeptical that they're actually capable of making something novel. There are thousands of hobby operating systems and video game emulators on github for it to train off of so it's not particularly surprising that it can copy somebody else's homework.

I remain confused but still somewhat interested as to a definition of "novel", given how often this idea is wielded in the AI context. How is everyone so good at identifying "novel"? For example, I can't wrap my head around how a) a human could come up with a piece of writing that inarguably reads "novel" writing, while b) an AI could be guaranteed to not be able to do the same, under the same standard.

Because we know that the human only read, say, fifty books since they were born, and watched a few thousand videos, and there is nothing in them which resembles what they wrote.

Re: A new Google model is nearly perfect on automated handwriting recognition

#136

Post by kazinator »

brulard wrote: post #88
snickerbockers wrote: post #76

Earlier quoted context omitted.

Generally novel either refers to something that is new, or a certain type of literature. If the AI is generating something functionally equivalent to a program in its training set (in this case, dozens or even hundreds of such programs) then it by definition cannot be novel.

This is quite a narrow view of how the generation works. AI can extrapolate from the training set and explore new directions. It's not just cutting pieces and gluing together.

Positively not. It is pure interpolation and not extrapolation. The training set is vast and supports an even vaster set of possible traversal paths; but they are all interpolative.

Same with diffusion and everything else. It is not extrapolation that you can transfer the style of Van Gogh onto a photographl it is interpolation.

Extrapolation might be something like inventing a style: how did Van Gogh do that?

And, sure, the thing can invent a new style---as a mashup of existing styles. Give me a Picasso-like take on Van Gogh and apply it to this image ...

Maybe the original thing there is the idea of doing that; but that came from me! The execution of it is just interpolation.

Re: A new Google model is nearly perfect on automated handwriting recognition

#137

Post by fooker »

neom wrote: post #87
HDThoreaun wrote: post #82

Earlier quoted context omitted.

It doesnt have to be perfect to be useful. If it does a decent job then your wife reviews and edits, that will be much faster than doing the whole thing by hand. The only question is if she can stay committed to perfection. I dont see the downside of trying it unless she's worried about getting lazy.

I raised this point with her, she said there are times it would be ambiguous for both her and the model, and she thinks it would be dangerous for her to be influenced by it. I'm not a professional historical researcher so I'm not sure if her concern is valid or not.

As a scientist, I don't think this is valid or useful. It's very much a first year PhD line of thought that academia stamps out of you.

This is the 'RE' in research, you specifically want to know and understand what others think of something by reading others' papers. The scientific training slowly, laboriously prepares you to reason about something without being too influenced by it.

Re: A new Google model is nearly perfect on automated handwriting recognition

#138

Post by magicalist »

mikestorrent wrote: post #114
testaccount28 wrote: post #42

Earlier quoted context omitted.

why would you admit on the internet that you fail the reverse turing test?

Didn't some fake AI country song just get on the top 100? How novel is novel? A lot of human artists aren't producing anything _novel_.

> Didn't some fake AI country song just get on the top 100?

No

Edit: to be less snarky, it topped the Billboard Country Digital Song Sales Chart, which is a measure of sales of the individual song, not streaming listens. It's estimated it takes a few thousand sales to top that particular chart and it's widely believed to be commonly manipulated by coordinated purchases.

Re: A new Google model is nearly perfect on automated handwriting recognition

#139

Post by CamperBob2 »

HarHarVeryFunny wrote: post #130
CamperBob2 wrote: post #90

Earlier quoted context omitted.

How'd you do at the International Math Olympiad this year?

I hear the LLM was able to parrot fragments of the stuff it was trained to memorize, and did very well

Yeah, that must be it.

Re: A new Google model is nearly perfect on automated handwriting recognition

#140

Post by MagicMoonlight »

It seems like a leap to assume it has done all sorts of complex calculations implicitly.

I looked at the image and immediately noticed that it is written as “14 5” in the original text. It doesn’t require calculation to guess that it might be 14 pounds 5 ounces rather than 145. Especially since presumably, that notation was used elsewhere in the document.

Post reply on HN