News for hackers. Stuff that matters.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @08:33AM (#38204759)
> We designed a test for this theory: we reran the benchmarks without showing the models the instructions to each exercise. Instead, we just told them that they were Exercism exercises, and gave them the exercise names and function stubs.

This summarizes all my skepticism agains the AI field. Pretty clear that they aren't solving the problem, they have them memorized.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by og_kalu on Thursday November 09, 2023 @08:35AM (#38204781)
"They memorized all the problems" is not what was found here and still a wrong overcorrection.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by dartos on Thursday November 09, 2023 @08:40AM (#38204822)
“Gpt-4 has more problems memorized than gpt-4 turbo” was exactly what was found here.

That doesn’t mean it’s only able to solve problems in its training set (tho it’s much better at that obviously.)

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @08:52AM (#38204974)
If you are shown only the title of a coding problem and the site name where it's from, and you manage to solve it you are showing that you either cheated or knew the answer.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by og_kalu on Thursday November 09, 2023 @08:54AM (#38204989)
I mean sure, it memorized some of the answers. I'm not denying that. Clearly, it didn't memorize all of them.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @08:57AM (#38205021)
When people say "oh look how amazing, it can solve programming problems!" when in fact it has only seen the models CHEAT, is an enormous problem.

For cases where finding the answer it's perfectly fine, but it's not fine for claims that it can code. There's a huge difference.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by notnaut on Thursday November 09, 2023 @09:10AM (#38205184)
It can generate never-before-seen strings of comprehensible language. It can react to the inherent logic embedded in words and text and provide a brute forced version of what a human could. That it can “solve” a problem only through “cheating” is an anthropomorphism that betrays the magic that is evident to anyone who has used these things.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by raincole on Thursday November 09, 2023 @09:08AM (#38205157)
Almost 2024 and people still can't accept that LLM can code...
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by danielmarkbruce on Thursday November 09, 2023 @11:10AM (#38206836)
Of course they can't. And self-driving cars also don't exist, it's like 10 years away at best.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by FeepingCreature on Thursday November 09, 2023 @09:00AM (#38205052)
I've seen it code on completely novel tasks, so I'm not sure what you're suggesting here. The model can unquestionably code.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by og_kalu on Thursday November 09, 2023 @09:02AM (#38205089)
Okay... Funny how forcing it to not CHEAT did not increase apparent ability.

It can code and it has memorized some coding questions are not mutually exclusive.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @09:11AM (#38205194)
> Okay... Funny how forcing it to not CHEAT did not increase apparent ability.

The article did the opposite. It forced the models to cheat to solve the problems. Which it did happily. It should have stated "there is no actual problem to solve here, you must supply a problem for me to solve".

> It can code and it has memorized some coding questions are not mutually exclusive

This I will give you. Many humans try to cheat at basic math because they are lazy, so will this model. Maybe that's a sign of intelligence :P

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by raincole on Thursday November 09, 2023 @09:26AM (#38205376)
Me: What's 6x6?

You: 36

Me: You cheated! You just cited the answer you memorized! You should have started from addition.

You: ...okay? 6+6=12, 12+6=18, 18+...

Me: You cheated again! You just have 6+6=12 memorized! You should make the rule of addition out of Peano axioms.

You: ...you're being annoying, but okay? First axiom, we define 0 as...

Me: You cheated again! You memorized Peano Axioms! Jesus Christ, is there any intelligent creature left?

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @09:48AM (#38205666)
But in this case it's not like that at all. They only saw the NAME of the problem. Like if I said "Page 23 of Mathbook Y, problem number 3". Which happens to be 6x6.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by furyofantares on Thursday November 09, 2023 @10:25AM (#38206164)
It said they gave the exercise name, which doesn't sound like just the exercise number but probably mildly descriptive -- and they also gave it function stubs.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by raincole on Thursday November 09, 2023 @09:55AM (#38205745)
Me: I was being in such a blah blah situation... does the article 3 of Digital Government Act applies here?

My lawyer: Hmm the article 3 says--

Me: I knew it! Lawyers are not intelligent...

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by stnmtn on Thursday November 09, 2023 @10:45AM (#38206460)
If I gave you a programming problem and all I told you was that the problem name was Traveling Salesman, you might be able to solve it based on that.

If not that, then if I just said "fizzbuzz" to you, I'm sure you would be able to give the solution without me needing to say any other descriptions of the problem

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @02:44PM (#38210059)
Again, because of memorization, not being able to code.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by FeepingCreature on Thursday November 09, 2023 @09:50PM (#38214512)
But in that case, not memorization of the specific problem set, but "programming background knowledge." Hardly something to blame the machine for when we rely on it every day.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by qup on Thursday November 09, 2023 @10:50AM (#38206532)
I know this is deep down a bad comment thread, but I thought I'd chime in here.

I have been writing function names and test names, and then telling gpt to fill in the test, which is usually does how I want (maybe with errors, but it tests the correct thing), and then I tell it to fill out the answers.

this is in a thing I'm building that's never been built, with names that I made up (but describe the functionality well)

It cannot have this spot memorized, I just invented it myself

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by jsight on Thursday November 09, 2023 @11:00AM (#38206658)
TBH, people underestimate how much of coding is just memorization. I'm guessing those of us with bad memories understand this more than the ones with good memories. :)

I can't remember how many times I've googled, "how do I create a directory in Python?". Now bard often generates an inline answer for me.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by hanselot on Thursday November 09, 2023 @09:26AM (#38205374)
Though this is exactly what happened. The initial test was ran on a model that "Cheated" (aka has memorized the answers). The second test was run on a model that didn't "Cheat" as much, yet still got only 2% less score. So, the question is not resolved really. How much did the first model cheat, and how much did the second? If the second model "cheats" less, then it wins.

Also, I don't understand your obsession with the word cheating. If you have solved a problem before on a different website and solve it again, did you cheat? Or did you just use your brain to store the solution for later?

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @09:40AM (#38205559)
> Also, I don't understand your obsession with the word cheating.

It's all about the rule set yea. Since the rule set is not defined, technically nothing is cheating. I just interpret the rule set as "can it code?" and for this rule set, it seems to me that it's cheating.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by boxed on Thursday November 09, 2023 @09:48AM (#38205679)
> How much did the first model cheat, and how much did the second? If the second model "cheats" less, then it wins.

They both cheated 100%. Because they both never saw the problem. AT ALL. They just saw the title and the name of the website.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by Terretta on Thursday November 09, 2023 @09:54AM (#38205738)
On the contrary, it could mean you were, to some percentage of success, able to guess what problem is, and then, to some multiplier percentage of success, solve it.

The key is, can you guess the problem from the title and the function name? I'd argue, sure, at least half the time?, why not...

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by mavhc on Thursday November 09, 2023 @08:35AM (#38204780)
LLMs are lossy compression
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by stoicfungi on Thursday November 09, 2023 @08:53AM (#38204978)
All models are, including humain brain.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by ChatGTP on Thursday November 09, 2023 @10:33AM (#38206293)
The human brain is a model?
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by rocgf on Thursday November 09, 2023 @10:42AM (#38206413)
It models the world around it, so it's fairly similar to what GPT does, especially with the newly-added image capabilities and stuff.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by ChatGTP on Thursday November 09, 2023 @10:34PM (#38214772)
But the brain itself is not a model.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by FrustratedMonky on Thursday November 09, 2023 @11:08AM (#38206805)
Consciousness itself is a model of the world.

Our experience of the world is a model executing.

Comparing the latest neuroscience to latest neural networks. They look and behave very similarly.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by DecayingOrganic on Thursday November 09, 2023 @08:57AM (#38205018)
Memorization often gets a bad rap as the underachiever's shortcut. However, it's a fundamental component of any learning process! Our ability to reason, solve problems, and innovate is all built upon a foundation of memorized information. In fact, it's precisely the reason humans have thrived for so long; we were able to memorize and pass down knowledge culturally long before the written word, not because we were 100 times smarter than our nearest cousins. Without solid memory, be it in our brains or AI algorithms, there's no foundation to build upon for higher reasoning.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by viraptor on Thursday November 09, 2023 @09:08AM (#38205153)
It's hard to decide for me without seeing the data. Even if you don't know the exact exercise, seeing the title and the function name/parameters is often enough for me to guess what the challenge is. I checked the public questions on exercism and almost all of those (that I spot checked) that contained the function name were extremely obvious. Knowing it's a programming challenge would also improve my guessing chances.

For example the function stubs I can find are "value_of_card(<card>)" in exercise "Black Jack", or "generate_seat_letters(<number>)" in exercise "Plane Tickets". I think I could guess those without seeing the rest of the question.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by GaggiX on Thursday November 09, 2023 @09:13AM (#38205214)
So how can it solve novel problems? Internet does not have all combinations for every possible task with any random programming language, library or constraints. It can even solve problems with non-existing programming languages and libraries if you describe them, if that's just memorization then I don't know what it isn't.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by m3kw9 on Thursday November 09, 2023 @09:30AM (#38205419)
From a black box point of view and one angle, gpt is a web filter where it will try to find you the exact thing you are looking for but from memory. Vs google you have to distill all the info into what you need
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by ren_engineer on Thursday November 09, 2023 @11:14AM (#38206918)
even if it's not true AI or even an architecture with the potential to become AI, LLMs are already good enough to provide real world value. Obviously "super autocomplete" isn't as sexy as true AI, but still very useful
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by drcode on Thursday November 09, 2023 @11:46AM (#38207362)
You can call it whatever you want, all I know is I used to write programs in lines of code, then blocks of code at a time, spit out by LLMs

Using GPT-4 Turbo yesterday, I feel like I'm moving to pages of code at a time now.

Taking the ideas in my head and turning them into reality is so easy now

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by blovescoffee on Thursday November 09, 2023 @12:07PM (#38207670)
Ok but you understand there's a body of literature that shows that LLMs don't "just" memorize
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by gloosx on Thursday November 09, 2023 @12:22PM (#38207917)
+100 to that. My biggest scepticism is people actually creating a new problem while thinking they are solving problem. Don't get me wrong, translating natural language ideas into code is fun and all, the truth it is also code, yet in ambiguous language format given to the machine.

When did natural language became better for expressing development ideas than code? I know – when you don't know how to code in the first place. Then you will have to bet on all of the ambiguities of the language, cultural and meta-physical which words carry in order to hack your thing together instead of expressing yourself directly and explicitly.

Finally what is beautiful about strict code format we are so used to - it is truly the fastest and shortest path to get your thing done, in case you possess the knowledge needed.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by ericrallen on Thursday November 09, 2023 @02:03PM (#38209371)
That sounds a lot like gatekeeping.

These tools will empower folks who aren’t developers to build stuff and maybe learn a bit more about how programming works.

They will enable folks who have ideas, but can’t express them, to actually be able to create what they are imagining.

That’s awesome.

Code isn’t beautiful (except for a few rare exceptions). Creating something with code is.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by gloosx on Friday November 10, 2023 @01:29AM (#38215752)
I agree it is a great tool for learning, but I don't believe anything more complex or of real use can be made AND maintained with it.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by ericrallen on Friday November 10, 2023 @03:28AM (#38216489)
I think we’re probably way to early in the AI lifecycle to really form any strongly held beliefs yet.

In the 11 months since ChatGPT was released, things have come a long way. Who knows where we’ll be in another 11 months.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by gloosx on Monday November 13, 2023 @05:24AM (#38248767)
What I'm trying to say is that the problem is not approachable this way at all – efficiently generating code by describing what you want, since when you compress what you want into a prompt you lose the details, and in order to restore all of them you will need a much bigger prompt volume than code generated. Because it is code itself which compresses an idea but no idea can compress the code well enough. In another 11 months it will be exactly in the same spot - it will not be able to be more efficient at this task by the nature of it.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by doug_durham on Thursday November 09, 2023 @03:20PM (#38210646)
Natural language isn't superior to computer languages. NL allows you to describe a software concept in a computer language and framework neutral way. The LLM generates the code. The real benefit is when you work across languages and frameworks. It is difficult to keep all of the details of all of the framework calls in your head all of the time.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by gloosx on Friday November 10, 2023 @01:39AM (#38215816)
>NL allows to describe a software concept in a computer language

What does that even mean? Natural language allows to describe in computer language? This is nonsense. The concept is only superficial, and you develop all the details of it using code itself. To outsource this development for a black-box which can get 60% of the tasks correctly is to learn a lot, but to loose all of the details.

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by gloosx on Friday November 10, 2023 @01:57AM (#38215937)
Where is the evidence for that? Any real-world application made and running by describing software concepts to an LLM?

It is what it is – a novel search engine, lossy and non-credible. Effectively useless on codebases that extend beyond its fairly limited context

Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by caesil on Thursday November 09, 2023 @12:31PM (#38208030)
If that's your takeaway from this then you really missed the point. The implication here is that gpt-4 to gpt-4-turbo represents a leap away from memorization and toward better reasoning with a more complete world model.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by nathanfig on Thursday November 09, 2023 @12:35PM (#38208081)
"memorize" implies they can only recite things verbatim and that's ignoring the massive leap in being able to synthesize disjoint "memories" in new ways.
Re:Benchmarking GPT-4 Turbo – A Cautionary Tale by rldjbpin on Friday November 10, 2023 @04:19AM (#38216782)
if the benchmark means replicating the experience of taking technical interviews by most people, then this is a spot on approach and serves the potential user right.