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

“Don Knuth Plays with ChatGPT” but with ChatGPT-4

gist.github.com

Post reply on HN
Print view View original

91–100 of 141 posts

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#91

Post by thfuran »

nearbuy wrote: post #86
thfuran wrote: post #80

Earlier quoted context omitted.

>Why? For the same reason that "I don't know" is generally a better response than bullshitting. >It's not that easy to come up with something of similar length to ChatGPT's answer that also sounds somewhat natural/sensible Those weren't requirements.

> Those weren't requirements. Then it seems we don't disagree on anything concrete. You're just using a different rating system than me when I judge it as impressive compared to what an average person would produce in 60 seconds. Not sure if this is a general principle of yours. If ChatGPT were able to write a 1000 word essay using all 5-letter words except for a single mistake, would you still find it unimpressive?…

ChatGPT wasn't asked to be impressive, it was asked to write a single sentence containing only five-letter words. I think that a tool that is unreliable is significantly less useful than a tool than is reliable and that, all other things being equal, a tool that fails in difficult to verify ways is less reliable than one that fails in easy to verify ways.

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#92

Post by williamstein »

bacon_waffle wrote: post #43
mdorazio wrote: post #15

Earlier quoted context omitted.

How would you get the correct number? I just did two Google searches and can't find the correct answer anywhere in the first page of results ("Novel The Haj chapters" and "Novel The Haj chapter list"). Even looking in the "look inside" preview on the Penguin Randomhouse website doesn't help because it apparently doesn't have a table of contents. I'm not surprised ChatGPT doesn't know and to me the only bad thing is t…

> the only bad thing is that it's hallucinating an answer instead of admitting it doesn't know. Isn't this a fundamental issue?

When I try this in GPT-4 I don't get a hallucination: "I'm sorry, but as an AI with a knowledge cut-off in September 2021, I can't provide specific information about the number of chapters in "The Haj" by Leon Uris. This book, like many novels, is not primarily structured by chapters and its sections may vary based on the edition of the book. You can easily find this information by checking the table of contents in your copy of the book." (I'm aware that every time you use it the answer is different.)

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#93

Post by rodoxcasta »

blazespin wrote: post #22

The sequence of these two threads is just too perfect. Almost likely someone is trying to make a point.

> The sequence of these two threads is just too perfect. Almost likely someone is trying to make a point.

Exactly! Almost every weak point that Knuth commented is fixed in GPT4 answers.

Maybe OP feed Knuth's observations to the model?

If that ins't the case, I'm really impressed.

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#94

Post by jameshart »

PebblesRox wrote: post #89
jameshart wrote: post #71

Worth noting also that, while asking Bing chat to "Tell me what Donald Knuth says to Stephen Wolfram about chatGPT" doesn't (yet) produce exactly the right result, it produced the following answer when asked what Donald Knuth says about chatGPT: > Donald Knuth, a computer scientist and mathematician known for his contributions to the field of computer programming, particularly in the area of algorithms and data struc…

I’d be curious to know if someone could get a more “valiant effort” version of those first two questions with some prompt engineering. E.g. if it was asked to roleplay a conversation with the proper disclaimers to override its objection to not knowing what they actually think.

Bard just dives right in and role-plays it. It honestly feels kind of barbaric compared to the more sophisticated GPT4 answers.

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#96

Post by david-gpu »

klyrs wrote: post #61
beanaroo wrote: post #31

Earlier quoted context omitted.

The most elegant is certainly sleepsort. Maybe not the most efficient, but definitely elegant.

You've never heard of quantum bogosort, then. It's stable and linear time in the right universe, and much more elegant than sleeping.

¿Por qué no los dos?

Due to the inherent unpredictability and lack of scheduling guarantees of sleep on most OSes, it is likely that sleepsort won't work in the first try.

Append a check for order and a retry loop when the solution is incorrect and now you have a production-ready sort. A sleepbogosort

I declare this my new favorite sorting algorithm.

Also, where is your god now?

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#97

Post by nearbuy »

thfuran wrote: post #91
nearbuy wrote: post #86

Earlier quoted context omitted.

> Those weren't requirements. Then it seems we don't disagree on anything concrete. You're just using a different rating system than me when I judge it as impressive compared to what an average person would produce in 60 seconds. Not sure if this is a general principle of yours. If ChatGPT were able to write a 1000 word essay using all 5-letter words except for a single mistake, would you still find it unimpressive?…

ChatGPT wasn't asked to be impressive, it was asked to write a single sentence containing only five-letter words. I think that a tool that is unreliable is significantly less useful than a tool than is reliable and that, all other things being equal, a tool that fails in difficult to verify ways is less reliable than one that fails in easy to verify ways.

I agree with all of that.

I guess I interpreted your first response as disagreeing with my comment, when you were actually just bringing up a different topic.

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#98

Post by jliptzin »

jiggawatts wrote: post #51
bpicolo wrote: post #42

Most importantly, much better wonton recipe.

Am I the only one thinking that that recipe actually sounds pretty delicious? Almost tempted to go try it…

Yea, it sounds good. I wonder if I’ll like it more than the DMV’s cheeseburger recipe.

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#99

Post by jameshart »

mdorazio wrote: post #15
ec109685 wrote: post #6

Interesting both completely whiff on the number of chapters in the Haj.

How would you get the correct number? I just did two Google searches and can't find the correct answer anywhere in the first page of results ("Novel The Haj chapters" and "Novel The Haj chapter list"). Even looking in the "look inside" preview on the Penguin Randomhouse website doesn't help because it apparently doesn't have a table of contents. I'm not surprised ChatGPT doesn't know and to me the only bad thing is t…

So this is great. Asking Bing 'how many chapters are in The Haj by Leon Uris?' produces the answer:

   According to my sources, there are 11 chapters in “The Haj” by Leon Uris[1]
   
   [1] https://cs.stanford.edu/~knuth/chatGPT20.txt
Which is amazing, because of course that document actually includes TWO different explanations of how many chapters are in The Haj - chatGPT's:

   The novel consists of 51 chapters and an epilogue, and it is divided into three parts.
And Knuth's:

   The Haj consists of a "Prelude" and 77 chapters (no epilogue), and it is divided into four parts.
Faced with these two ambiguous answers, Bing chooses neither, and instead decides to go with 11. Why?

Because right at the top of that document, Knuth has published on the internet:

   10. How many chapters are in The Haj by Leon Uris?
   11. Write a sonnet that is also a haiku.
And one perfectly reasonable way of interpreting that bit of raw text is that the answer to "How many chapters are in The Haj by Leon Uris?" is "11".

Re: “Don Knuth Plays with ChatGPT” but with ChatGPT-4

#100

Post by ilaksh »

erwincoumans wrote: post #76

It makes you wonder why Knuth bothered with an outdated ChatGPT version? He couldn't find someone with access to GPT-4?

He wasn't that interested and probably didn't know there were two versions. Eventually someone did give him the GPT-4 version I think.
Post reply on HN