Ok but calculators are only allowed in math class and if there’s one thing language models suck at, it’s doing basic math. Forget anything at least as complicated as algebra
For me they weren’t allowed in Calc I, II, III, Alg I, II and Differential equations. Every other class pretty much required it.
if there’s one thing language models suck at, it’s doing basic math.
If you’re using a GPT 3.5 turbo level models, sure. Synthetic data is perfect for teaching LLMs, o1 will be good enough up to Calc III IMO, maybe even better.
The only thing I don’t like about this is that it uses a TI, yikes.
They let us use them for all my college math classes.
They really don’t help much at all if you don’t understand the math, and if you do understand, you don’t need the calculator most of the time.
That’s also what my upper level math courses were like in college. In high school and the first couple of years of college I got good use out of my graphing calculator, but after that I reached the point where all of its advanced features were no longer useful. I just ended up leaving it at home and brought my old TI-30 Solar for class for the occasional time I had to crunch some actual numbers.
Don’t know about university math, but this applied to a lot of the stuff in my last years of school. Since we always had a part where you were required to solve everything without a calculator you had to be able to do everything without it. For algebra and Calculus it just meant that you were able to do the math more efficiently. For statistics the calculator was basically useless, since it didnt help you if you didnt knew what you had to do, what was basically the only hard part of it.
LLMs do suck at math, if you look into it, the o1 models actually escape the LLM output and write a python function to calculate the output, I’ve been able to break their math functions by asking for functions that use math not in the standard Python library.
I know someone also wrote a wolfram integration to help solve LLMs math problems.
Wow that’s really clever actually. Basically using the library as digital scratch paper
Terrence Tao (one of the most famous and active mathematician) recently wrote his thoughts in Mastodon on o1 mathematical capabilities. Interesting read: https://mathstodon.xyz/@tao/113132502735585408
Thanks for sharing, knew him from some numberphile vids cool to see they have a mastadon account. Good to know that LLMs are crawling from “incompentent graduate” to “mediocre graduate”. Which basically means its already smarter than most people for many kinds of reasoning task.
I’m not a big fan of the way the guy speaks though, as is common for super intelligent academic types they have to use overly complicated wording to formally describe even the most basic opinions while mixing in hints of inflated ego and intellectual superiority. He should start experimenting with having o-1 as his editor and summarize his toots.
The language wasn’t that complex
TI, yikes.
Yeah, well, TI has spent bucketloads of money bribing textbook publishers to only include instruction for their specific models so they are now the de facto standard in American schools. This is apparently legal.
Anyway, team Casio represent.
Sorry, I’m an HP guy. I love their calculators, hate everything else they do, except their plotters maybe.
They’re great at multiple choice when they’ve seen the test versions
Time to build a wolfarm applet
Dude, I was wondering why someone hadn’t done this the moment they discovered Ai was terrible at math. I would have imagine the crowds who deal with both would have some overlap at least.
This kid never took science
You have an option not to take science?
Damn. Where I live the three main sciences were mandatory and they were all separate subjects
Better than plastic explosives and shrapnel.
Oof getting ChatGPT to help on a test is likely to lead to some wrong answers.
deleted by creator
You know what they say: C’s get degrees.
“C’s get degrees, and here’s the tease: no one’s asking for transcripts, just expertise.”
I’ve had to provide my transcript for several job applications. Government ones seem to usually ask.
10 years in?
My senior dev was asked for his transcript and he’s been in the field for 25+years. He told them to get in touch with his college that doesn’t exist anymore. Suspect it’s a standard set of forms they need to fill.
It would have to be a unicorn position to even begin to consider that.
Idk, if there is one thing it does consistently well its standardized tests.
Not that it can be used in any non mathematical class and if teachers do actually pay attention.
Cept you can’t use any wifi models on any standardized test for this reason.
It’s not a WiFi model, a custom module was hidden inside the calculator to provide the WiFi connection.
Ah yup should’ve read the article lol. That’s a whole lot of work and effort into cheating, which probably won’t work? Needs a whole thing to it sounds like plug into the link port? Which would stick out… so like… idk MAYBE they are stupid and don’t notice but like… I wouldn’t bet my life on that.
Because the phrase “Cheaters never prosper” isn’t actually true. There are many executives, politicians and rich people that very very likely “cheated” at some point along the way to get where they are, multiple times probably.
Hell, I wouldn’t even be surprised if such phrases were spread by the rich/ruling class/whatever as just another thing to keep people divided lol
There’s a video in the article showing the whole process. The new module was completely hidden inside the calculators case and soldered to the internal connections.
Until you actually open it up, it doesn’t look abnormal at all.
Ah, interesting. That’s quite a bit of work more than expected then lol. Link things at the top assumed it’d plug into that. Seems like a whole lot of work where if you’re into fixing shit and soldering and all that you probably don’t hate math much and thus… learning algebra 2 for a damn SAT wouldn’t be that hard?
I suppose maybe the GRE or whatever? Idk what all that entails and if they allow these.
Wat? Does it have wifi?
They added wifi with a extra circuit board hidden inside the calculator case. It’s connected to the calculators communication port, and pretends to be another calculator. So they can use the calculator’s built in “send” function to send variables/text/etc to the hidden card, which then uses it’s internet connection to look up answers and send the results back.
Yep
If a kid is smart enough to figure this out and make it work for them, they’re gonna be fine…
As someone who was a kid who would do things like this to avoid putting in the work, no this kid will probably not be fine.
Yes, but the kids buying the modded devices may not be
good. they will learn not to buy their way out of a problem at least.
Back when we were doing quadratic equations; I wrote a program on my TI-84 that would ask which parts of the equation you already had, and would fill in the rest for you.
My teacher liked it so much he bought a transfer cable for those calculators so he could get a copy for himself. Then used to to grade tests.
I made one to decompose polynomials it was very good because it showed all the steps it was literally just copy what’s on the calc to the page
So you didn’t get the transfer cable with your calculator? Smells fishy
Issued by the school; I never owned it.
you can code directly on the device, it’s just a PAIN to do compared to moving the files over
I could never remember the formula to calculate compound interest.
But I had no trouble writing a for loop.
What always annoyed me was having to draw charts by hand. Just let me put the data in a computer for god’s sake, the rest of the working is there… I did actually write a python function for one of my assignments which was fine, but they told me not to do it for the exam.
K•(1+r)^n
I would just rebuild something in my head like this every time.
While i < n; k=k+(k*r); i++;
You’d think I could remember k(1+r)^n but when you posted, it looked as alien as it felt decades ago.
The use of for makes sense.
k=0; for (i=0; i<n; i++) k=k+f(i);is the same ask=\sum_{i=0}^{n-1} f(i)and
k=1; for (i=0; i<n; i++) k=k*f(i);is the same ask=\prod_{i=0}^{n-1} f(i)In our case,
f(i)=1+randk=1; for (i=0; i<n; i++) k*(1+r);is the same ask=\prod_{i=0}^{n-1} (1+r) = (1+r)^nAll of that just to say that exponentiation is an iteration of multiplication, the same way that multiplication is an iteration of addition
I did the same thing. It was allowed in general, with the correct thought, “if you can code it yourself, you know the content”
I had another “program” that would fail to run but that’s because I wrote notes into it. Doubt that was allowed.
Here in NZ they do a factory reset on your calculator at the start of every exam.
Oh I would have been so pissed. I was programming on my calculator 24/7 instead of my classes.
I wrote a sudoku “editor”
I put that in quotes because I had a grid that could be navigated, arrows moved, storing the numbers, had number entry down, and then I learned the hard way what p vs np is.
They did that here too, but students would use a cheat program that made it look like teachers were resetting it, but really the memory was safe
I don’t remember if they fully closed the loopholes, but there are inputs that programs cannot catch unless you actually replace the OS.
I did that but made it return success before it got to the notes. You had to scroll to get to the notes, but it looked innocuous before that.
But can you still play Drug Wars?
Has anyone ever beaten that game? Is it even possible?
It’s possible to get to the last day and have a lot of money, which is your “score.” The world leaderboards have some ridiculous highscores on the remake called Dope Wars.
https://stevekola.tripod.com/dopewarsguide/id2.html
That is a strategy guide.
I have never managed to do it.
I played whatever version was easily accessible on Ubuntu’s repository in 2009, and once managed to get a ton of bitches and effectively infinite money. I have no idea how I did it.
The earliest exponention clicker game I ever discovered. Lost dozens of hours to ti83 drug wars
No one has beaten The Game, but it is possible. We just need to wait for the Pope to declare that it has come to an end
Oh you fucker, its been over ten years. i just lost The Game.
Game has been over for about that long. https://xkcd.com/391/
Sounds fake but okay.
Saw the video. Quite cool mod.
https://youtu.be/Bicjxl4EcJgNot sure if it’s the same thing but a few days ago I saw a youtube video where person modded it with a wifi card so it could communicate with your pc which is at home. It required internet access from your phone which needs to be near though.
Not anymore since it’s spreading news instead of remaining on YouTube
I wonder what can counter this except banning it, or provide calculators to students instead of using their own.
Utilizing the tools available to you to solve problems is not cheating, its resourcefulness, and using your brain. Which is of course frowned upon in schools that exist to churn out mindless drones for corporate enslavement.
ChatGPT is famously bad at the things you’d use a calculator for though
Obligatory plug for calcgpt
Id see it potentially being useful for starting a problem. Could probably point you towards the right direction for physics problem even if the numbers are off
But you also have a calculator…
Its been quite a while since I’ve taken a proctored exam, but then all the proctors would clear all the memory on your calc before they’d let you use it for test. Is that not the case anymore?
The launcher program can be downloaded on-demand, avoiding detection if a teacher inspects or clears the calculator’s memory
If I understood it correctly, the Wi-Fi module appears as a standard calculator-to-calculator interface, so built-in commands can install the cheat apps at any time.
Depends on the exam. Some don’t even allow programmable calcs because they don’t want to deal with possible shit like this. I have already seen a certification exam where they provide the calculators as well.
The article said it can be download “on demand” so that might make the clearing pointless.
Make people switch devices, problem solved. Does not work without tampering with the hardware
What would happen if now plug in another calculator? AFAIK that only a P2P connection and never meant for 2 parties.
Bring your calculator to the Spanish exam. Trust me, this plan is flawless.
This is cool stuff
Hot stuff. I got to say, YouTube has some pretty interesting things.
“ChatGPT what is the formula for Work Done in an enclosed system expressed as a triple integral?”
“42”
“Ok cool ty.”
Stop giving me Thermo nightmares; I lived through that shit already I don’t need to sleep through it too.
Yeah but at least work on an enclosed system is always zero. Idk why but I always chuckle about that.
Sure, you can prove it in like 4 to 8 lines of multivariate calculus, but its always gonna be 0.
