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

Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

arxiv.org

Post reply on HN
Print view View original

51–60 of 62 posts

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#51

Post by mftb »

dmreedy wrote: post #24
mftb wrote: post #17

Earlier quoted context omitted.

Yea it's not. Sorry to contradict, but it's not like that. In any kind of tutoring arrangement you're time with them is limited, and if they're any good, they don't just regurgitate limitless example code. Two of the most important decisions that an instructor has to make are, how much access to give you, and how much example material to give you, because the actual learning begins when you have to think for yourself…

Tutoring assumes the skill is valuable to learn, that there is a need for more people who know how to do it. We don't really tutor people how to write too much assembly anymore, or hand-compile code. So if you're arguing that ChatGPT meets the definition of a tool, or a servant, better than a tutor, fair, but if you're further arguing that that makes it somehow less valuable than a tutor (in this case), I'm not sure…

Yea, I definitely wasn't trying to quantify it's value. ChatGPT definitely appears to be proving valuable to people. I was just challenging the idea that so far it's acting in a tutor/instructor/mentor type of role. While it seems like an interesting direction to take these LLMs, so far I haven't observed them doing that.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#52

Post by jillesvangurp »

People aren't that good at logic either. So, gpt-4 not being great at this is maybe not that surprising.

Probably the best feature of gpt-4 is the ability to use tools. For example, it may not be that good at calculating things. But it can use a calculator. And if you think about it, a lot of people (including mathematicians) aren't actually that good at calculating either. We all learn it in school and then we forget much of it. That's why we have calculators. It's not a big deal.

Gpt-4 is more than capable of knowing the best tool for the job. Figuring out how to use it isn't that hard. You can actually ask it "what's the best tool for X", get a usable response, and then ask a follow up question to produce a script in the language of your choosing that demonstrates how to use it, complete with unit tests. Not a hypothetical, I've been doing that in the past few weeks and I've been getting some usable results.

And that's put me in a mind of wondering what will happen once we start networking all these specialized AIs and tools together. It might not be able to do everything by itself but it can get quite far figuring out requirements and turning those into running code. It's not that big of a leap from answering questions about how to do things to actually building programs that do things.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#53

Post by Closi »

jltsiren wrote: post #12
TOMDM wrote: post #6

Earlier quoted context omitted.

I think a lot of these LLM benchmarks should include a human avg, otherwise I don't really have a frame of reference other than personal experience with the models.

Human average can be a misleading statistic, because the average human is useless for almost everything. In almost every job, the average person doing the job is well above the average (in the general population) for that particular job.

I think this just demonstrates how the goalposts are shifting though.

Until pretty recently most people would probably say “the average human is very flexible at solving reasoning tasks compared to machines which find reasoning incredibly challenging“.

Now it’s “well of course this AI which wasn’t specifically trained for verbal reasoning can beat an average human at verbal reasoning - humans are useless at almost everything!”

Your goalpost seems to be that GPT needs to be better than experts in their field to be considered “good” at something - but I think it’s just interesting to reflect that that’s the benchmark we are applying now.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#54

Post by bakuninsbart »

svachalek wrote: post #19
Closi wrote: post #3

I think one main failure in the framing of these papers (and discussion of LLMs more broadly) is that the abstract says that GPT4 ‘struggles’ with logical reasoning: > ChatGPT and GPT-4 do relatively well on well-known datasets […] however, the performance drops significantly when handling newly released and out-of-distribution [where] Logical reasoning remains challenging for ChatGPT and GPT-4 But reading the paper…

If you really force it to reason, rather than regurgitate arguments from its training set, you will find it is nowhere near the genius line. Make up some rules and have it try to answer questions according to the rules. In my experiments I feel it's something like a 4 or 5 year old child both in its logical limitations and penchant for distraction. However it's important to note one VERY important thing -- this is no…

Agreed, but let's not forget that Carnap started his AI company last year with the express goal of reaching AGI comparable to a 'retarded toddler' by 2030. Relatively simple generative AIs have come far far further than anyone really anticipated, and it is quite unclear if the last 20% to avg-human-level AGI will be much harder, impossible, or also suddenly be solved. I mean, hell, GPT4s context space is still relatively small, it doesn't have a memory, and is still producing quite impressive results in simple reasoning tasks.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#55

Post by Closi »

skybrian wrote: post #15
Closi wrote: post #3

I think one main failure in the framing of these papers (and discussion of LLMs more broadly) is that the abstract says that GPT4 ‘struggles’ with logical reasoning: > ChatGPT and GPT-4 do relatively well on well-known datasets […] however, the performance drops significantly when handling newly released and out-of-distribution [where] Logical reasoning remains challenging for ChatGPT and GPT-4 But reading the paper…

You're framing this as if there were a single yes-or-no question that we should all agree on. (Are the LLM's "any good?") But in real-world contexts, there are some tasks that just about anyone could do, others where "average" human performance isn't good enough and you need to hire an expert, and also some jobs that can only be done by machine. So it seems like the bar should be set based on what you think is necess…

Not a simple 'yes-or-no' question, but more about the framing and where the benchmark is.

When they conclude that GPT4 "does not perform astonishingly well" - what is this compared to?

They never define what 'doing well' looks like, were not able to identify an application that does better than GPT4, and also were not able to say what a human benchmark would be if given the same task.

I can say though that I read the sample question and got it wrong too, so these aren't trivial questions we are giving GPT4.

So based on this, I just don't really understand how they can support their conclusion that it "does not perform astonishingly well".

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#56

Post by causality0 »

cjbprime wrote: post #50
causality0 wrote: post #20

They're good in "memory" reasoning but terrible in deductive reasoning. Like if you say there's a sign in front of a door saying "push" it will tell you you need to push the door, but if you say there was a powerful wind and you see a sign saying "pull" laying on the ground on the other side of a glass door it has no idea if you should push or pull.

I guess I'm with the LLM on this one, since I can't follow your example. Did the sign flip over while it was falling? Did the sign fall towards or away from the glass door that I am on the other side of? Where are the doorhandles? Can you write this example in a way that's more comprehensible to humans, and then we can ask GPT-4 about it?

The sign only has one side. It's a sign saying "pull" that was knocked off the other side of a glass door by the wind.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#57

Post by seba_dos1 »

dbrueck wrote: post #48
seba_dos1 wrote: post #47

Earlier quoted context omitted.

> I'm talking about a mental model for understanding and predicting when and why it works well vs when it doesn't. I'm talking about it too. If I enter a specific phrase into a search engine that can be only found on a handful of websites, I expect it to return those results to me. Like, typing the VAT ID of my company will return bunch of information about it on various sites. This is absolutely not going to work wi…

Again, we seem to be talking past each other, sorry. I'm really, really, really not talking about the search engine itself. I'm talking about the hunk of tech that makes up the interface layer between the human and the search engine, and the fact that that hunk of tech can be hooked up to a search engine is interesting but not entirely germane. If using the analogy of human memory works for you - that's great! To me,…

> I'm talking about a mental model for understanding and predicting when and why it works well vs when it doesn't.

That's what you said earlier you were talking about, and that's what I replied to. Now you're saying that you're in fact not talking about "the mental model of understanding what works with LLMs and doesn't" at all. Seems you have to improve your communication skills mate ;]

What I'm saying is that using LLMs while imagining them to be kinda like search engines is just a way to get burned by hallucinations and disappointed with poor results. They don't work even remotely similar to search engines, neither internally nor for an external observer. For some kinds of input they may trick you into believing they actually do, but that impression will fall apart pretty quickly once you try to actually exercise it. That's how you get people who are genuinely shocked that ChatGPT gave them references to papers that were completely made up, for example - which is something that shouldn't surprise anyone using this tech at all, as that's just how it works.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#58

Post by anothernewdude »

pottspotts wrote: post #7
RayVR wrote: post #2

This aligns well with my personal experience using gpt-4. The model provides surprisingly good responses on topics which I know are readily available online while being potentially troublesome to find the exact information I want. I have even found it useful when I know there is a tool for what I want but can’t recall the jargon used to find it via Google. Simply describing the rough idea is enough to get the model t…

For a significant number of software developers, GPT and Github's Copilot have replaced StackOverflow, and even Googling more generally. It is more than an autocomplete, it is the best resource for software development by far, IMO. It's a tutor that's an expert in virtually every topic.

Expert as of 2021, which is obsolete for many software dev purposes, not that SO is much better.

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#59

Post by skybrian »

Closi wrote: post #55
skybrian wrote: post #15

Earlier quoted context omitted.

You're framing this as if there were a single yes-or-no question that we should all agree on. (Are the LLM's "any good?") But in real-world contexts, there are some tasks that just about anyone could do, others where "average" human performance isn't good enough and you need to hire an expert, and also some jobs that can only be done by machine. So it seems like the bar should be set based on what you think is necess…

Not a simple 'yes-or-no' question, but more about the framing and where the benchmark is. When they conclude that GPT4 "does not perform astonishingly well" - what is this compared to? They never define what 'doing well' looks like, were not able to identify an application that does better than GPT4, and also were not able to say what a human benchmark would be if given the same task. I can say though that I read the…

You’re right that they don’t compare to people at all, and the benchmarks don’t show performance on a practical application. And I agree that the last sentence isn’t great, but I don’t think it’s that important. I guess they were hoping it would do better on the benchmarks? It’s not an objective statement.

You don’t read a paper for its conclusion. A good question to ask about a scientific paper is “what did they actually do?” In this case, they asked ChatGPT (presumably GPT3.5) and GPT4 a bunch of logical reasoning questions from some benchmarks and compared the benchmark scores to RoBERTa. That’s it. Running benchmarks can be useful, but how much you care about the benchmarks is up to you.

Higher scores are better, so it does seem promising that GPT4 got more questions right. The scores aren’t that meaningful me, but it seems like it’s objective confirmation that GPT4 is better than previous systems on logical reasoning?

Maybe the benchmark scores are more meaningful to someone else? What else have these benchmarks been used for?

Re: Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4

#60

Post by Closi »

skybrian wrote: post #59
Closi wrote: post #55

Earlier quoted context omitted.

Not a simple 'yes-or-no' question, but more about the framing and where the benchmark is. When they conclude that GPT4 "does not perform astonishingly well" - what is this compared to? They never define what 'doing well' looks like, were not able to identify an application that does better than GPT4, and also were not able to say what a human benchmark would be if given the same task. I can say though that I read the…

You’re right that they don’t compare to people at all, and the benchmarks don’t show performance on a practical application. And I agree that the last sentence isn’t great, but I don’t think it’s that important. I guess they were hoping it would do better on the benchmarks? It’s not an objective statement. You don’t read a paper for its conclusion. A good question to ask about a scientific paper is “what did they act…

I think we are probably just evaluating the paper on different metrics too :)

I think my view is just that if your paper is called "Evaluating the Logical Reasoning Ability of GPT-4" and your conclusion is "logical reasoning remains challenging for GPT4" then you should have something in your paper to back up that statement that's more objective, particularly if the findings appear to be that it performs better at logical reasoning than anything else the paper identifies to date.

It's supposed to be an academic paper, not a tumblr post.

Post reply on HN