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

The emoji problem (2022)

artofproblemsolving.com

Post reply on HN
Print view View original

51–60 of 71 posts

Re: The emoji problem (2022)

#51

Post by lblume »

I tried giving this to ChatGPT. (Just by uploading the image to the base OpenAI interface.) I expected to either (a) have the model already know the question and give the right answer, (b) hallucinate an answer or (c) refuse to engage with the problem at all.

Instead, this happened:

https://chatgpt.com/share/682cce62-c53c-8003-be2c-2929395868...

Basically, the model confidently outputs a guess, then calculates it, determines it to be incorrect, and repeatedly tries again, even repeating the same guesses over and over. It does not recognize any symmetry and acts like a completely unstructured agent. In the end, the model vehemently asserts there to be no solutions to this puzzle. I really did not expect this and will update my beliefs accordingly if the models behave as badly with future puzzles.

Re: The emoji problem (2022)

#52

Post by ToucanLoucan »

behnamoh wrote: post #36
stronglikedan wrote: post #29

Earlier quoted context omitted.

now that's peak Quora right there!

I really miss those years when Quora was a decent place. it went downhill when they decided to aggressively monetize it. the same happened to poe.

I heard they use LLMs to generate questions and answers to those questions to help drive traffic, which is why no matter what insane bullshit you type into google, you will probably get a quora page about it.

Something something dead internet theory.

Re: The emoji problem (2022)

#53

Post by robinhouston »

pas wrote: post #50
robinhouston wrote: post #27

Using other constants in place of the ‘4’ can lead to some _really_ gigantic smallest solutions: https://observablehq.com/@robinhouston/a-remarkable-diophant...

Ah, I really really love the dumb but magnificent on-demand loading of the 120-million-digit-long numbers in a nice little textarea!

I had fun with that! I'm happy someone appreciates it.

S3's support of the Range: header makes it possible to do that sort of thing remarkably cheaply and efficiently.

Re: The emoji problem (2022)

#54

Post by somat »

adzm wrote: post #24

On a related tangent, when I was teaching my younger kids about math and helping with homework, I would often rewrite things as a formula, or rewrite the formula itself once they got to that point. However instead of things like x I would use things like fluffy cloud, star, etc. They thought it was annoying but it still kept them interested and have said they've done the same when helping their peers. It's easy to fo…

> it could be the phrase "total number of kittens"

There is this minimalism present in math culture, while I sort of understand it, when tossing and mixing formulas it helps to have everything as small as possible. but later, when it is published it really sucks for readability, "Okay here is an item doing some heavy lifting in this formula what is it for? hell if I know some joker labeled it 'φ' "

I like to joke, If you think programmers are bad at naming thing you should see the mathematicians, they take a perverse pride in their inability to name things.

The worst are programs derived directly from a math paper, if your variable holds the correlation coefficient call it that. we have thousands of years of language and labels we can use to share our ideas with others. don't encrypt it and call it "rho".

Re: The emoji problem (2022)

#55

Post by oytis »

qingcharles wrote: post #43
oytis wrote: post #4

It's year 2025, why doesn't the author use actual fruit emoji for variable names?

Here's Gemini's (brute force?) solution to the problem in C# using fruit emoji variables: https://imgur.com/a/cC5QPH0

Nice! Not going to work, but appreciate the emoji.

UPD. This code is like that picture with Disney princesses. The closer you look, the scarier it gets

Re: The emoji problem (2022)

#56

Post by LeoPanthera »

lblume wrote: post #51

I tried giving this to ChatGPT. (Just by uploading the image to the base OpenAI interface.) I expected to either (a) have the model already know the question and give the right answer, (b) hallucinate an answer or (c) refuse to engage with the problem at all. Instead, this happened: https://chatgpt.com/share/682cce62-c53c-8003-be2c-2929395868... Basically, the model confidently outputs a guess, then calculates it, de…

I'm impressed. There certainly are no "reasonable" solutions, where I am arbitrarily defining reasonable as numbers that humans can work or understand in their heads.

Isn't there a version of ChatGPT that connects to Wolfram Alpha? Did you try that one?

Re: The emoji problem (2022)

#57

Post by celticninja »

ysofunny wrote: post #40
celticninja wrote: post #19

Earlier quoted context omitted.

i think based on the username they seem to be looking for validation through upvotes. Perhaps assuming this place is like a nerdier reddit.

actually, reddit is a mainstream hacker news to me old.reddit.com will always be reddit.com

Ah I never changed from old.reddit, I am sure the current version is entirely unrecognisable. It throws me sometimes when a comment mentions a users pfp, because wtf? Reddit doesn't have user profiles with pictures.

Re: The emoji problem (2022)

#58

Post by morkalork »

ToucanLoucan wrote: post #52
behnamoh wrote: post #36

Earlier quoted context omitted.

I really miss those years when Quora was a decent place. it went downhill when they decided to aggressively monetize it. the same happened to poe.

I heard they use LLMs to generate questions and answers to those questions to help drive traffic, which is why no matter what insane bullshit you type into google, you will probably get a quora page about it. Something something dead internet theory.

I saw a reddit thread from sub about {specific retailer} in {specific geographical region} where the OP ends with ChatGPT's classic "Would you like some more variations of this text?" response and the whole damn thread was other users happily replying on topic, totally missing that OP was clearly a bot because they were probably bots too. But hey, they all get karma, all get activity and look they're members of {geographic region} so their opinions on future political threads in {region} are totally legit right?

But yeah, something something dead internet theory indeed.

Re: The emoji problem (2022)

#59

Post by JdeBP »

somat wrote: post #54
adzm wrote: post #24

On a related tangent, when I was teaching my younger kids about math and helping with homework, I would often rewrite things as a formula, or rewrite the formula itself once they got to that point. However instead of things like x I would use things like fluffy cloud, star, etc. They thought it was annoying but it still kept them interested and have said they've done the same when helping their peers. It's easy to fo…

> it could be the phrase "total number of kittens" There is this minimalism present in math culture, while I sort of understand it, when tossing and mixing formulas it helps to have everything as small as possible. but later, when it is published it really sucks for readability, "Okay here is an item doing some heavy lifting in this formula what is it for? hell if I know some joker labeled it 'φ' " I like to joke, If…

One "fun" aspect to this is that there's not just 1 but 13 extra copies of the English alphabet in Unicode that are there specifically for mathematicians to name stuff in formulae and equations.

And with the exceptions of Fraktur and script, free fonts that are already sans-serif and monospace generally do not make the glyphs visually distinct from the actual English alphabet characters similarly weighted and slanted.

Which gives the sighted people a taste of what the sight-impaired people have to suffer when people abuse these code points as ways to write in italics or boldface, because the screen readers correctly obey the important note in Unicode Technical Report 25 that these letters do not combine to make words, and read them out as mathematical formulae where "d" is multiplied by "o" and then by "g".

Re: The emoji problem (2022)

#60

Post by JdeBP »

cout wrote: post #45
socalgal2 wrote: post #25

Earlier quoted context omitted.

Good or not, many languages choose not to support emoji, even newer ones. The reasons are complicated but, many people prefer unicode normalization so that different forms of what appear to be the same word are considered the same word. People argue whether or not this is important but it can certainly be argued that it would be frustrating to get an error like let café = 1; café += 1; // error, unknown identifier '…

I had an idea the other day to use 64-bit character codes. Then each code can be interpreted as an 8x8 bitmap. Every character would be guaranteed to have a unique bitmap representation. The bitmaps wouldn't bet used for rendering, of course -- but they could be used as a fallback if your font does not define a character. Anyway this would somewhat avoid the problem you describe because two characters that look the s…

As someone who has worked with 8×8 fonts I can report that you'd have some surprising problems with that idea. Not only would you have problems with there being two distinct forms of letters like "a" and "g", making things over-unique; and not only is it tricky to differentiate forms that actually are not the same, because they are from two different alphabets (especially the 13 extra "mathematical" alphabets); but it's actually quite difficult to make pre-composed forms in that amount of space.

8×8 is a tight squeeze, and 16×16 works a lot better. But that would make your approach vastly more space hungry than a normalization approach using the actual Unicode code points.

* https://github.com/jdebp/unscii/tree/2.1.1f

Post reply on HN