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

Learning to Reason with LLMs

openai.com

Post reply on HN
Print view View original

511–520 of 1001 posts

Re: Learning to Reason with LLMs

#511

Post by mindcrime »

AIorNot wrote: post #467
LouisSayers wrote: post #431

Earlier quoted context omitted.

Don't do it, help us keep our high salaries :D Joking aside, even with AI generating code, someone has to know how to talk to it, how to understand the output, and know what to do with it. AI is also not great for novel concepts and may not fully get what's happening when a bug occurs. Remember, it's just a tool at the end of the day.

just change this to "I have AI Skills!!" :) https://www.youtube.com/watch?v=hNuu9CpdjIo

Not having clicked the link yet, I'm going to speculate that this is the famous Office Space "I have people skills, damnit!" scene.

...

And it was. :-) Nice callback!

Re: Learning to Reason with LLMs

#512

Post by lukev »

This is a pretty big technical achievement, and I am excited to see this type of advancement in the field.

However, I am very worried about the utility of this tool given that it (like all LLMs) is still prone to hallucination. Exactly who is it for?

If you're enough of an expert to critically judge the output, you're probably just as well off doing the reasoning yourself. If you're not capable of evaluating the output, you risk relying on completely wrong answers.

For example, I just asked it to evaluate an algorithm I'm working on to optimize database join ordering. Early in the reasoning process it confidently and incorrectly stated that "join costs are usually symmetrical" and then later steps incorporated that, trying to get me to "simplify" my algorithm by using an undirected graph instead of a directed one as the internal data structure.

If you're familiar with database optimization, you'll know that this is... very wrong. But otherwise, the line of reasoning was cogent and compelling.

I worry it would lead me astray, if it confidently relied on a fact that I wasn't able to immediately recognize was incorrect.

Re: Learning to Reason with LLMs

#513

Post by reliabilityguy »

jazzyjackson wrote: post #454
w4 wrote: post #439

Earlier quoted context omitted.

There are two basic versions of “safety” which are related, but distinct: One version of “safety” is a pernicious censorship impulse shared by many modern intellectuals, some of whom are in tech. They believe that they alone are capable of safely engaging with the world of ideas to determine what is true, and thus feel strongly that information and speech ought to be censored to prevent the rabble from engaging in wr…

Third version is "brand safety" which is, we don't want to be in a new york times feature about 13 year olds following anarchist-cookbook instructions from our flagship product

Do you think that 13 year olds today can’t find this book on their own?

Re: Learning to Reason with LLMs

#514

Post by zaptheimpaler »

cs391231 wrote: post #408

Student here. Can someone give me one reason why I should continue in software engineering that isn't denial and hopium?

I agree there's too much cope going around. All the people saying AI is just a tool to augment our jobs are correct, humans are still needed but perhaps far less of them will be needed. If job openings shrink by 50% or disproportionately impact juniors it will hurt.

One decent reason to continue is that pretty much all white collar professions will be impacted by this. I think it's a big enough number that the powers that be will have to roll it out slowly, figure out UBI or something because if all of us are thrown into unemployment in a short time there will be riots. Like on a scale of all the jobs that AI can replace, there are many jobs that are easier to replace than software so its comparatively still a better option than most. But overall I'm getting progressively more worried as well.

Re: Learning to Reason with LLMs

#515

Post by packetlost »

hidelooktropic wrote: post #430
packetlost wrote: post #177

lol at the graphs at the top. Logarithmic scaling for test/compute time should make everyone who thinks AGI is possible with this architecture take pause.

I don't see any log scaled graphs.

The two first graphs on the page are labelled as log scale in the time axis, so I don't know what you're looking at but it's definitely there.

Re: Learning to Reason with LLMs

#516

Post by lopatamd »

sterlind wrote: post #506
cs391231 wrote: post #408

Student here. Can someone give me one reason why I should continue in software engineering that isn't denial and hopium?

Because none of your other majors will hold up much longer. Once software engineering becomes fully automated, so will EE, ME, applied math, economics, physics, etc. If you work with your hands, like a surgeon or chemist, you'll last longer, but the thinky bits of those jobs will disappear. And once AI research is automated, how long will it be until we have dexterous robots? So basically, switching majors is just ru…

[dead]

Re: Learning to Reason with LLMs

#518

Post by fwip »

bartman wrote: post #367

This is incredible. In April I used the standard GPT-4 model via ChatGPT to help me reverse engineer the binary bluetooth protocol used by my kitchen fan to integrate it into Home Assistant. It was helpful in a rubber duck way, but could not determine the pattern used to transmit the remaining runtime of the fan in a certain mode. Initial prompt here [0] I pasted the same prompt into o1-preview and o1-mini and both c…

What's the incredible part here? Being able to write code to turn hex into decimal?

Re: Learning to Reason with LLMs

#519

Post by fzaninotto »

fzaninotto wrote: post #438

It can solve sudoku. It took 119s to solve this easy grid: _ 7 8 4 1 _ _ _ 9 5 _ 1 _ 2 _ 4 7 _ _ 2 9 _ 6 _ _ _ _ _ 3 _ _ _ 7 6 9 4 _ 4 5 3 _ _ 8 1 _ _ _ _ _ _ _ 3 _ _ 9 _ 4 6 7 2 1 3 _ 6 _ _ _ _ _ 7 _ 8 _ _ _ 8 3 1 _ _ _

It seems to be unable to solve hard sudokus, like the following one where it gave 2 wrong answers before abandoning.

+-------+-------+-------+ | 6 . . | 9 1 . | . . . | | 2 . 5 | . . . | 1 . 7 | | . 3 . | . 2 7 | 5 . . | +-------+-------+-------+ | 3 . 4 | . . 1 | . 2 . | | . 6 . | 3 . . | . . . | | . . 9 | . 5 . | . 7 . | +-------+-------+-------+ | . . . | 7 . . | 2 1 . | | . . . | . 9 . | 7 . 4 | | 4 . . | . . . | 6 8 5 | +-------+-------+-------+

So we're safe for another few months.

Re: Learning to Reason with LLMs

#520

Post by thruway516 »

refulgentis wrote: post #459
cs391231 wrote: post #408

Student here. Can someone give me one reason why I should continue in software engineering that isn't denial and hopium?

I went from economics dropout waiter who built a app startup with $0 funding and $1M a year in revenue by midway through year 1, sold it a few years later, then went to Google for 7 years, and last year I left. I'm mentioning that because the following sounds darn opinionated and brusque without the context I've capital-S seen a variety of people and situations. Sit down and be really honest with yourself. If your go…

Lol. I like this answer. You can either think of it in terms of "it'll eat my lunch" or "I now have 10x more capabilities and can be 100x more productive". The former category will be self-fulfilling.
Post reply on HN