1281
you are viewing a single comment's thread
view the rest of the comments
[-] Thorry84@feddit.nl 141 points 6 months ago
[-] douglasg14b@lemmy.world 35 points 6 months ago* (last edited 6 months ago)

Generative AI is INCREDIBLY bad at mathmatical/logical reasoning. This is well known, and very much not surprising.

That's actually one of the milestones on the way to general artificial intelligence. The ability to reason about logic & math is a huge increase in AI capability.

[-] callcc@lemmy.world 4 points 6 months ago

Well known by you, not everybody.

[-] fallingcats@discuss.tchncs.de 4 points 6 months ago* (last edited 6 months ago)

Well known by everyone that knows anything about LLMs at all

[-] kromem@lemmy.world 0 points 6 months ago* (last edited 6 months ago)
[-] fallingcats@discuss.tchncs.de 5 points 6 months ago

I've used gpt4 enough in the past months to confidently say the improvements in this blog post aren't noteworthy

[-] kromem@lemmy.world 1 points 6 months ago

They aren't live in the consumer model. This is a research post, not in production.

There's other literature elsewhere on getting improved math performance with GPT-4 as it exists right now.

[-] kromem@lemmy.world 0 points 6 months ago

It's really not in the most current models.

And it's already at present incredibly advanced in research.

The bigger issue is abstract reasoning that necessitates nonlinear representations - things like Sodoku, where exploring a solution requires updating the conditions and pursuing multiple paths to a solution. This can be achieved with multiple calls, but doing it in a single process is currently a fool's errand and likely will be until a shift to future architectures.

[-] douglasg14b@lemmy.world 1 points 6 months ago

I'm referring to models that understand language and semantics, such as LLMs.

Other models that are specifically trained can't do what it can, but they can perform math.

[-] kromem@lemmy.world 1 points 6 months ago

The linked research is about LLMs. The opening of the abstract of the paper:

In recent years, large language models have greatly improved in their ability to perform complex multi-step reasoning. However, even state-of-the-art models still regularly produce logical mistakes. To train more reliable models, we can turn either to outcome supervision, which provides feedback for a final result, or process supervision, which provides feedback for each intermediate reasoning step. Given the importance of training reliable models, and given the high cost of human feedback, it is important to carefully compare the both methods. Recent work has already begun this comparison, but many questions still remain. We conduct our own investigation, finding that process supervision significantly outperforms outcome supervision for training models to solve problems from the challenging MATH dataset. Our process-supervised model solves 78% of problems from a representative subset of the MATH test set. Additionally, we show that active learning significantly improves the efficacy of process supervision.

[-] Trollception@lemmy.world 15 points 6 months ago

So that's correct... Or am I dumber than the AI?

[-] JGrffn@lemmy.world 92 points 6 months ago

If one gallon is 3.785 liters, then one gallon is less than 4 liters. So, 4 liters should've been the answer.

[-] Smc87@lemmy.sdf.org 84 points 6 months ago
[-] WhiteHawk@lemmy.world 43 points 6 months ago
[-] Matty_r@programming.dev 19 points 6 months ago

4l is only 2 characters, 3.785l is 6 characters. 6 > 2, therefore 3.785l is greater than 4l.

[-] Klear@sh.itjust.works 7 points 6 months ago* (last edited 6 months ago)

You're forgetting the decimal point. The second one is just 1.4 characters.

[-] intensely_human@lemm.ee 2 points 6 months ago

“4” > “3.785”

=> false

[-] nifty@lemmy.world 1 points 6 months ago

That’s maybe how GPT reasoned it as well, you could be an LLM whisperer.

[-] stolid_agnostic@lemmy.ml 19 points 6 months ago

Everyone has a bad day now and then so don’t worry about it.

[-] fossphi@lemm.ee 2 points 6 months ago

Ummm... username check out?

[-] moog@lemm.ee 0 points 6 months ago

U are dumber than the AI ig lol

[-] SomeoneSomewhere@lemmy.nz 3 points 6 months ago

Obviously it's referring to the 4.54609 litre UK gallon /s

[-] kromem@lemmy.world 1 points 6 months ago

You can see from the green icon that it's GPT-3.5.

GPT-3.5 really is best described as simply "convincing autocomplete."

It isn't until GPT-4 that there were compelling reasoning capabilities including rudimentary spatial awareness (I suspect in part from being a multimodal model).

In fact, it was the jump from a nonsense answer regarding a "stack these items" prompt from 3.5 to a very well structured answer in 4 that blew a lot of minds at Microsoft.

this post was submitted on 27 Dec 2023
1281 points (95.9% liked)

Microblog Memes

4771 readers
2308 users here now

A place to share screenshots of Microblog posts, whether from Mastodon, tumblr, ~~Twitter~~ X, KBin, Threads or elsewhere.

Created as an evolution of White People Twitter and other tweet-capture subreddits.

Rules:

  1. Please put at least one word relevant to the post in the post title.
  2. Be nice.
  3. No advertising, brand promotion or guerilla marketing.

Related communities:

founded 1 year ago
MODERATORS