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

Irrelevant facts about cats added to math problems increase LLM errors by 300%

science.org

Post reply on HN
Print view View original

261–270 of 270 posts

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#261

Post by ants_everywhere »

littlestymaar wrote: post #251
ants_everywhere wrote: post #247

Earlier quoted context omitted.

[flagged]

Well, your statements show that you are much less informed than you believe.

As I've said, go look at the literature.

If you find in the literature any evidence for your theory that the people who invented these concepts believed they had no relationship to the biological concepts then please contribute it.

If you're unwilling to do the bare minimum required to take the opposite of my position then you aren't really defending a point of view you're just gainsaying reflexively.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#262

Post by littlestymaar »

ants_everywhere wrote: post #261
littlestymaar wrote: post #251

Earlier quoted context omitted.

Well, your statements show that you are much less informed than you believe.

As I've said, go look at the literature. If you find in the literature any evidence for your theory that the people who invented these concepts believed they had no relationship to the biological concepts then please contribute it. If you're unwilling to do the bare minimum required to take the opposite of my position then you aren't really defending a point of view you're just gainsaying reflexively.

Flexing about literature you haven't read is poor taste, especially since you are misrepresenting it.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#263

Post by ants_everywhere »

littlestymaar wrote: post #262
ants_everywhere wrote: post #261

Earlier quoted context omitted.

As I've said, go look at the literature. If you find in the literature any evidence for your theory that the people who invented these concepts believed they had no relationship to the biological concepts then please contribute it. If you're unwilling to do the bare minimum required to take the opposite of my position then you aren't really defending a point of view you're just gainsaying reflexively.

Flexing about literature you haven't read is poor taste, especially since you are misrepresenting it.

I have read it, I'm not the least bit misrepresenting it.

Since you're just resorting to deliberately lying about things I don't see a reason to pursue this further.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#264

Post by squidbeak »

mjburgess wrote: post #220
squidbeak wrote: post #217

Earlier quoted context omitted.

> It's also off-the-charts implausible to say that our performance on adding up substantially degrades with the introduction of irrelevant information Didn't you ever sit an exam next to a irresistibly gorgeous girl? Or haven't you ever gone to work in the middle of a personal crisis? Or filled out a form while people were rowing in your office? Or written code with a pneumatic drill and banging outdoors? That's the…

"Intelligence" is a metaphor used to describe LLMs (, AI) used by those who have never studied intelligence. If you had studied intelligence as a science of systems which are intelligent (ie., animals, people, etc.) then this comparison would seem absurd to you; mendacious and designed to confound. The desperation to find some scenario in which, at the most extreme superficial level, an intelligent agent "benchmarks…

You seem to be replying to a completely different post. You'll see I didn't once use the term 'intelligence', so the reprimand you lead with about the use of that term is pretty odd.

The ramble that follows has its curiosities, not least the compulsion you have to demean or insult your 'gullible', 'credulous' opponents, but is otherwise far from any point. The contention of yours I was replying to was your curiously absolute statement that human performance doesn't degrade with the introduction of irrelevant information. I gave you instances any of us can relate to where it definitely does degrade. Rather than dispute my point, you've allowed some kind of 'extra information' to bounce you around irrelevancies from one tangent to the next - through torture, blenders, animals as systems, etc etc. What you've actually done, quite beautifully, is restate my point for me.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#265

Post by Terretta »

Y_Y wrote: post #3

> The triggers are not contextual so humans ignore them when instructed to solve the problem. Do they? I've found humans to be quite poor at ignoring irrelevant information, even when it isn't about cats. I would have insisted on a human control group to compare the results with.

If you spell “sit in the tub” s-o-a-k soak, and you spell “a funny story” j-o-k-e joke, how do you spell “the white of an egg”?

Context engineering* has been around longer than we think. It works on humans too.

The cats are just adversarial context priming, same as this riddle.

* I've called it "context priming" for a couple years for reasons showed by this child's riddle, while considering "context engineering" as iteratively determining what priming unspools robust resilient results for the question.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#266

Post by n4r9 »

squidbeak wrote: post #264
mjburgess wrote: post #220

Earlier quoted context omitted.

"Intelligence" is a metaphor used to describe LLMs (, AI) used by those who have never studied intelligence. If you had studied intelligence as a science of systems which are intelligent (ie., animals, people, etc.) then this comparison would seem absurd to you; mendacious and designed to confound. The desperation to find some scenario in which, at the most extreme superficial level, an intelligent agent "benchmarks…

You seem to be replying to a completely different post. You'll see I didn't once use the term 'intelligence', so the reprimand you lead with about the use of that term is pretty odd. The ramble that follows has its curiosities, not least the compulsion you have to demean or insult your 'gullible', 'credulous' opponents, but is otherwise far from any point. The contention of yours I was replying to was your curiously…

I may not agree with you but I appreciate your efforts to call out demaning and absolutist language on HN. It really drags the discussion down.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#267

Post by littlestymaar »

ants_everywhere wrote: post #263
littlestymaar wrote: post #262

Earlier quoted context omitted.

Flexing about literature you haven't read is poor taste, especially since you are misrepresenting it.

I have read it, I'm not the least bit misrepresenting it. Since you're just resorting to deliberately lying about things I don't see a reason to pursue this further.

Why would anyone read the original perceptron paper in this century though? It's easy to know someone is bullshitting when they claim to have read 70 years old papers that aren't in themselves of any interest nowadays. (Like how many economists are quoting David Ricardo without having read it directly, because reading Ricardo is a very poor way of spending your time and energy).

Funny how you spent lots of time in this thread talking shit to professionals of the fields and then take offence when someone call you like the fool you are.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#268

Post by beezlewax »

acc_297 wrote: post #80

There is more than one comment here asserting that the authors should have done a parallel comparison study against humans on the same question bank as if the study authors had set out to investigate whether humans or LLMs reason better in this situation. The authors do include the claim that humans would immediately disregard this information and maybe some would and some wouldn't that could be debated and seemingly…

They're already missed deployed in society

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#269

Post by Kuinox »

kazinator wrote: post #260
Kuinox wrote: post #240

Earlier quoted context omitted.

The problem with your answer is that you make affirmations using logical fallacies. We both don't know how LLMs, and brains works to produce output. Any affirmation toward that without proof is affirming things without any basis. For example in this response: > the brain can wield language without having to scan anywhere near hundreds of terabytes of text. The amount of text we need to train an LLM only goes down, ev…

Training the weights of the neural network produces a humungous function with a vast number of parameters. Such a function is not inherently mysterious due to the size alone. For instance, if we fit a billion numeric points to a polynomial curve having a billion coefficients, we would not be mystified as to how the polynomial interpolates between the points. Be that as it may, the trained neural network function does…

> we would not be mystified as to how the polynomial interpolates between the points.

We would not be surprised - we wouldnt know how the model resolve the problem. We wouldn't know if it is approximating, calculating the correct value, or memorising result. We would only know how it was built. We would be mystified in how it solved the problem.

> But that doesn't mean we don't know how it works. We invented it and produced it by training.

It is not because it was invented that we know how it works. The fallacy in your reasoning is thinking that Emergent Behavior or Properties can be trivialy explained, by knowing it's building block.

Re: Irrelevant facts about cats added to math problems increase LLM errors by 300%

#270

Post by mjburgess »

squidbeak wrote: post #264
mjburgess wrote: post #220

Earlier quoted context omitted.

"Intelligence" is a metaphor used to describe LLMs (, AI) used by those who have never studied intelligence. If you had studied intelligence as a science of systems which are intelligent (ie., animals, people, etc.) then this comparison would seem absurd to you; mendacious and designed to confound. The desperation to find some scenario in which, at the most extreme superficial level, an intelligent agent "benchmarks…

You seem to be replying to a completely different post. You'll see I didn't once use the term 'intelligence', so the reprimand you lead with about the use of that term is pretty odd. The ramble that follows has its curiosities, not least the compulsion you have to demean or insult your 'gullible', 'credulous' opponents, but is otherwise far from any point. The contention of yours I was replying to was your curiously…

So you strawman'd my claim about degradation of performance to one in which "substantial", "irrelevant" and "almost all cases" have no flexibility to circumscribe scenarios, so that i must be making a universal claim... And then you take issue with my reply?

Why would you think that I'd deny that you can't find scenarios in which performance substantially degrades? Would I not countenance toture? As in my reply?

My reply is against your presumption that an appropriate response to the spirit-and-plain-meaning of my argument is to "go and find another scenario". It is this presumption, when addressed, short-circuits this scenario-finding dialogue: In my reply I address the whole families of scenarios you are appealing to where we fail to function well and show why there existence remains irrelevant to our analysis of llms

Post reply on HN