[Haskell-cafe] GPT & Haskell

[Aloïs Cochard]

Absolutely no idea if there is an improvement, but this will should
eventually help for such mathematical shenanigans:
https://thenewstack.io/wolfram-chatgpt-plugin-blends-symbolic-ai-with-generative-ai/

Anyway, I hope everyone had a good April Fool's Day.
<https://en.wikipedia.org/wiki/April_Fools%27_Day>

On Sun, 2 Apr 2023 at 01:39, Mihai Maruseac <mihai.maruseac at gmail.com>
wrote:

> At the risk of touting my own horn, I have been running experiments with
> chatGPT a while ago and both math (https://mihai.page/chatgpt/) and
> Haskell (https://mihai.page/chatgpt-2/) were still not there.
>
> I need to repeat these experiments though for the new version, these were
> run when GPT-3 was top of the line.
>
> On Sat, Apr 1, 2023 at 3:38 PM Aloïs Cochard <alois.cochard at gmail.com>
> wrote:
>
>> I'm sorry, I'm too busy adding constraints to the next version of Chat
>> j'ai peter.
>>
>> On Sat, 1 Apr 2023 at 23:26, Branimir Maksimovic <
>> branimir.maksimovic at gmail.com> wrote:
>>
>>> Give it to solve this bellow one minute in Haskell:
>>> #512 Sums of totients of powers <https://projecteuler.net/problem=512>
>>> projecteuler.net <https://projecteuler.net/problem=512>
>>> [image: apple-touch-icon.png] <https://projecteuler.net/problem=512>
>>> <https://projecteuler.net/problem=512>
>>> I have shown it solution, but anyway really doubt it is programmed
>>> to remember anything :P
>>>
>>> Greets, Branimir.
>>>
>>> On 1. 4. 2023., at 18:52, Viktor Dukhovni <ietf-dane at dukhovni.org>
>>> wrote:
>>>
>>> On Sat, Apr 01, 2023 at 08:18:08AM +0200, Aloïs Cochard wrote:
>>>
>>> How can this be useful when you have to anyway review everything is doing
>>> as he might to just randomly insert a bug or a security flaw???
>>> I prefer to read poems by my human friends.
>>>
>>> I highly recommend starting reading this paper at page 128 instead of
>>> wasting your time on that prompt:
>>> https://arxiv.org/pdf/2303.12712.pdf
>>>
>>> Be ready for a good laughhi
>>>
>>>
>>> Part of the good laugh is on page 136:
>>>
>>>    Let P be a point on the hyperbola
>>>
>>>        x^2 + 3xy + 4x − 5y = −9y^2 − 133.
>>>
>>>    Find the shortest possible distance from the origin to P.
>>>
>>> The authors of the paper say that "GPT-4 produces a sound argument",
>>> I beg to differ.
>>>
>>>    Let u = 3y, then
>>>
>>>        x^2 + xu + u^2 + 4x − (5/3)u = − 133.
>>>
>>>    The degree-two part of which is positive-definite.  The linear terms
>>>    just shift the origin. So the equation is actually:
>>>
>>>        r^2 + rs + s^2 = RHS
>>>
>>>    For a straight-forward to compute choice of r = x - a, s = u - b.
>>>
>>> There are then two issues (just the first one is enough) with the prompt:
>>>
>>>    * The equation can't represent a hyperbola, it would be an ellipse.
>>>    * The ellipse doesn't exist, because the RHS constant is actually
>>> negative.
>>>
>>> THe authors are just as prone to autopilot nonsense reasoning as GPT-4.
>>>
>>> This rather reminds me of:
>>>
>>>
>>> https://math.stackexchange.com/questions/1594740/v-i-arnold-says-russian-students-cant-solve-this-problem-but-american-student
>>>
>>> --
>>>    Viktor.
>>> _______________________________________________
>>> Haskell-Cafe mailing list
>>> To (un)subscribe, modify options or view archives go to:
>>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
>>> Only members subscribed via the mailman list are allowed to post.
>>>
>>>
>>> _______________________________________________
>>> Haskell-Cafe mailing list
>>> To (un)subscribe, modify options or view archives go to:
>>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
>>> Only members subscribed via the mailman list are allowed to post.
>>
>>
>>
>> --
>> *Λ\oïs*
>> http://twitter.com/aloiscochard
>> http://github.com/aloiscochard
>> _______________________________________________
>> Haskell-Cafe mailing list
>> To (un)subscribe, modify options or view archives go to:
>> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
>> Only members subscribed via the mailman list are allowed to post.
>
>
>
> --
> Mihai Maruseac (MM)
> "If you can't solve a problem, then there's an easier problem you can
> solve: find it." -- George Polya
>


-- 
*Λ\oïs*
http://twitter.com/aloiscochard
http://github.com/aloiscochard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/haskell-cafe/attachments/20230402/621547ae/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: apple-touch-icon.png
Type: image/png
Size: 4501 bytes
Desc: not available
URL: <http://mail.haskell.org/pipermail/haskell-cafe/attachments/20230402/621547ae/attachment.png>