<div dir="ltr">At the risk of touting my own horn, I have been running experiments with chatGPT a while ago and both math (<a href="https://mihai.page/chatgpt/">https://mihai.page/chatgpt/</a>) and Haskell (<a href="https://mihai.page/chatgpt-2/">https://mihai.page/chatgpt-2/</a>) were still not there.<div><br></div><div>I need to repeat these experiments though for the new version, these were run when GPT-3 was top of the line.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sat, Apr 1, 2023 at 3:38 PM Aloïs Cochard <<a href="mailto:alois.cochard@gmail.com">alois.cochard@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">I'm sorry, I'm too busy adding constraints to the next version of Chat j'ai peter.<br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sat, 1 Apr 2023 at 23:26, Branimir Maksimovic <<a href="mailto:branimir.maksimovic@gmail.com" target="_blank">branimir.maksimovic@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>Give it to solve this bellow one minute in Haskell:<div><div style="display:block"><div style="display:inline-block" role="link"><a style="border-radius:10px;font-family:-apple-system,Helvetica,Arial,sans-serif;display:block;width:300px;overflow:hidden;text-decoration:none" rel="nofollow" href="https://projecteuler.net/problem=512" dir="ltr" role="button" width="300" target="_blank"><table style="table-layout:fixed;border-collapse:collapse;width:300px;background-color:rgb(229,230,233);font-family:-apple-system,Helvetica,Arial,sans-serif" width="300" cellspacing="0" cellpadding="0" border="0"><tbody><tr><td><table style="font-family:-apple-system,Helvetica,Arial,sans-serif;table-layout:fixed;background-color:rgb(229,230,233)" width="300" cellspacing="0" cellpadding="0" bgcolor="#E5E6E9"><tbody><tr><td style="padding:8px 0px"><div style="max-width:100%;margin:0px 16px;overflow:hidden"><div style="font-weight:500;font-size:12px;overflow:hidden;text-overflow:ellipsis;text-align:left"><a rel="nofollow" href="https://projecteuler.net/problem=512" style="text-decoration:none" target="_blank"><font style="color:rgba(0,0,0,0.847)" color="#272727">#512 Sums of totients of powers</font></a></div><div style="font-weight:400;font-size:11px;overflow:hidden;text-overflow:ellipsis;text-align:left"><a rel="nofollow" href="https://projecteuler.net/problem=512" style="text-decoration:none" target="_blank"><font style="color:rgba(0,0,0,0.498)" color="#808080">projecteuler.net</font></a></div></div></td><td style="padding:6px 12px 6px 0px" width="36"><a rel="nofollow" href="https://projecteuler.net/problem=512" target="_blank"><img style="display: inline-block; width: 36px; height: 36px; border-radius: 3px;" alt="apple-touch-icon.png" src="cid:1873ef82d308e3918d51" width="36" height="36"></a></td></tr></tbody></table></td></tr></tbody></table></a></div></div>I have shown it solution, but anyway really doubt it is programmed</div><div>to remember anything :P</div><div><br></div><div>Greets, Branimir.<br><div><br><blockquote type="cite"><div>On 1. 4. 2023., at 18:52, Viktor Dukhovni <<a href="mailto:ietf-dane@dukhovni.org" target="_blank">ietf-dane@dukhovni.org</a>> wrote:</div><br><div><div>On Sat, Apr 01, 2023 at 08:18:08AM +0200, Aloïs Cochard wrote:<br><br><blockquote type="cite">How can this be useful when you have to anyway review everything is doing<br>as he might to just randomly insert a bug or a security flaw???<br>I prefer to read poems by my human friends.<br><br>I highly recommend starting reading this paper at page 128 instead of<br>wasting your time on that prompt:<br><a href="https://arxiv.org/pdf/2303.12712.pdf" target="_blank">https://arxiv.org/pdf/2303.12712.pdf</a><br><br>Be ready for a good laughhi<br></blockquote><br>Part of the good laugh is on page 136:<br><br> Let P be a point on the hyperbola<br><br> x^2 + 3xy + 4x − 5y = −9y^2 − 133.<br><br> Find the shortest possible distance from the origin to P.<br><br>The authors of the paper say that "GPT-4 produces a sound argument",<br>I beg to differ.<br><br> Let u = 3y, then<br><br> x^2 + xu + u^2 + 4x − (5/3)u = − 133.<br><br> The degree-two part of which is positive-definite. The linear terms<br> just shift the origin. So the equation is actually:<br><br> r^2 + rs + s^2 = RHS<br><br> For a straight-forward to compute choice of r = x - a, s = u - b.<br><br>There are then two issues (just the first one is enough) with the prompt:<br><br> * The equation can't represent a hyperbola, it would be an ellipse.<br> * The ellipse doesn't exist, because the RHS constant is actually negative. <br><br>THe authors are just as prone to autopilot nonsense reasoning as GPT-4.<br><br>This rather reminds me of:<br><br> <a href="https://math.stackexchange.com/questions/1594740/v-i-arnold-says-russian-students-cant-solve-this-problem-but-american-student" target="_blank">https://math.stackexchange.com/questions/1594740/v-i-arnold-says-russian-students-cant-solve-this-problem-but-american-student</a><br><br>-- <br> Viktor.<br>_______________________________________________<br>Haskell-Cafe mailing list<br>To (un)subscribe, modify options or view archives go to:<br><a href="http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe" target="_blank">http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe</a><br>Only members subscribed via the mailman list are allowed to post.</div></div></blockquote></div><br></div></div>_______________________________________________<br>
Haskell-Cafe mailing list<br>
To (un)subscribe, modify options or view archives go to:<br>
<a href="http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe" rel="noreferrer" target="_blank">http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe</a><br>
Only members subscribed via the mailman list are allowed to post.</blockquote></div><br clear="all"><br><span>-- </span><br><div dir="ltr"><div dir="ltr"><div><div dir="ltr"><div><b>Λ\oïs</b></div><div><div><a href="http://twitter.com/aloiscochard" target="_blank">http://twitter.com/aloiscochard</a></div><div><a href="http://github.com/aloiscochard" target="_blank">http://github.com/aloiscochard</a></div></div></div></div></div></div>
_______________________________________________<br>
Haskell-Cafe mailing list<br>
To (un)subscribe, modify options or view archives go to:<br>
<a href="http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe" rel="noreferrer" target="_blank">http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe</a><br>
Only members subscribed via the mailman list are allowed to post.</blockquote></div><br clear="all"><div><br></div><span class="gmail_signature_prefix">-- </span><br><div dir="ltr" class="gmail_signature">Mihai Maruseac (MM)<br>"If you can't solve a problem, then there's an easier problem you can solve: find it." -- George Polya</div>