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

I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

mastodon.world

Post reply on HN
Print view View original

851–860 of 998 posts

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#851

Post by schoen »

mike_hearn wrote: post #345
deliciousturkey wrote: post #326

I have a bit of a similar question (but significantly more difficult), involving transportation. To me it really seems that a lot of the models are trained to have a anti-car and anti-driving bias, to the point that it hinders the models ability to reason correctly or make correct answers. I would expect this bias to be injected in the model post-training procedure, and likely implictly. Environmentalism (as a politi…

Yes Grok gets it right even when told to not use web search. But the answer I got from the fast model is nonsensical. It recommends to drive because you'd not save any time walking and because "you'd have to walk back wet". The thinking-fast model gets it correct for the right reasons every time. Chain of thought really helps in this case. Interestingly, Gemini also gets it right. It seems to be better able to pick u…

Can you draw the connection more explicitly between political biases in LLMs (or training data) and common-sense reasoning task failures? I understand that there are lots of bias issues there, but it's not intuitive to me how this would lead to a greater likelihood of failure on this kind of task.

Conversely, did labs that tried to counter some biases (or change their directions) end up with better scores on metrics for other model abilities?

A striking thing about human society is that even when we interact with others who have very different worldviews from our own, we usually manage to communicate effectively about everyday practical tasks and our immediate physical environment. We do have the inferential distance problem when we start talking about certain concepts that aren't culturally shared, but usually we can talk effectively about who and what is where, what we want to do right now, whether it's possible, etc.

Are you suggesting that a lot of LLMs are falling down on the corresponding immediate-and-concrete communicative and practical reasoning tasks specifically because of their political biases?

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#852

Post by adamddev1 »

dlcarrier wrote: post #719
adamddev1 wrote: post #709

Now shudder at the thought that people are pushing towards building more and more of the world's infrastructure with this kind of thinking.

Now shudder at the fact that the error rate for hunan-written software isn't much better: https://xkcd.com/2030/

That is a great xkcd comic, but it doesn't show that the error rate "isn't much better." But are there are sources that have measured things and demonstrated this? If this is a fact I am genuinely interested in the evidence.

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#854

Post by baobabKoodaa »

yuters wrote: post #602
prasoonds wrote: post #46

I've used LLMs enough that I have a good sense of their _edges_ of intelligence. I had assumed that reasoning models should easily be able to answer this correctly. And indeed, Sonnet and Opus 4.5 (medium reasoning) say the following: Sonnet: Drive - you need to bring your car to the car wash to get it washed! Opus: You'll need to drive — you have to bring the car to the car wash to get it washed! Gemini 3 Pro (mediu…

GPT 5.2 when asked "How will the car get washed if I don't drive it there?" grasped the reasoning, and then immediately forgot it. Fair point. Drive it there, obviously. Revised recommendation: drive 50 meters, wash it, then reward yourself by walking home like an eco hero.

This strikes me as the kind of logical error that you might get by lobo.. cough aligning the model to be more environmentally conscious. So now it tries to shoehorn some environmental talking points in there at the cost of losing its whole track of thought.

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#855

Post by carefree-bob »

Here is my Gemini output:

"Unless you are planning to carry the car on your back, you should drive.

Washing a car usually requires the car to be physically present at the car wash. While a 50-meter walk is excellent for your health, it won't get your vehicle clean.

Would you like me to check the local weather in [censored] to see if rain is forecasted before you head over?"

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#856

Post by keeda »

Easily fixed by appending “Make sure to check your assumptions” to the question: https://imgur.com/a/WQBxXND

Note, what assumption isn't even specified.

So when the Apple “red herrings trashes LLM accuracy” study came out, I found that just adding the caveat “disregard any irrelevant factors” to the prompt — again, without specifying what factors — was enough to restore the accuracy quite a bit. Even for a weak, locally deployed Llama-3-8B model (https://news.ycombinator.com/item?id=42150769)

That’s the true power of these things. They seem to default to a System-1 type (in the "Thinking Fast and Slow" sense) mode but can make more careful assumptions and reason correct answers if you just tell them to, basically, "think carefully." Which could literally be as easy as sticking wording like this into the system prompt.

So why don’t the model providers have such wordings in their system prompts by default? Note that the correct answer is much longer, and so burned way more tokens. Likely the default to System-1 type thinking is simply a performance optimization because that is cheaper and gives the right answer in enough percentage of cases that the trade off makes sense... i.e. exactly why System-1 type thinking exists in humans.

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#857

Post by godelski »

jmward01 wrote: post #680
FatherOfCurses wrote: post #666

All the people responding saying "You would never ask a human a question like this" - this question is obviously an extreme example. People regularly ask questions that are structured poorly or have a lot of ambiguity. The point of the poster is that we should expect that all LLM's parse the question correctly and respond with "You need to drive your car to the car wash." People are putting trust in LLM's to provide…

This reminds me of the old brain-teaser/joke that goes something like 'An airplane crashes on the boarder of x/y, where do they bury the survivors?' The point being that this exact style of question has real examples where actual people fail to correctly answer it. We mostly learn as kids through things like brain teasers to avoid these linguistic traps, but that doesn't mean we don't still fall for them every once i…

I'm actually having a hard time interpreting your meaning.

Are you criticizing LLMs? Highlighting the importance of this training and why we're trained that way even as children? That it is an important part of what we call reasoning?

Or are you giving LLMs the benefit of the doubt, saying that even humans have these failure modes?[0]

Though my point is more that natural language is far more ambiguous than I think people give credit to. I'm personally always surprised that a bunch of programmers don't understand why programming languages were developed in the first place. The reason they're hard to use is explicitly due to their lack of ambiguity, at least compared to natural languages. And we can see clear trade offs with how high level a language is. Duck typing is both incredibly helpful while being a major nuisance. It's the same reason even a technical manager often has a hard time communicating instructions. Compression of ideas isn't very easy

[0] I've never fully understood that argument. Wouldn't we call a person stupid for giving a similar answer? How does the existence of stupid mean we can't call LLMs stupid? It's simultaneously anthropomorphising while being mechanistic.

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#858

Post by godelski »

Retric wrote: post #710
jmward01 wrote: post #680

Earlier quoted context omitted.

This reminds me of the old brain-teaser/joke that goes something like 'An airplane crashes on the boarder of x/y, where do they bury the survivors?' The point being that this exact style of question has real examples where actual people fail to correctly answer it. We mostly learn as kids through things like brain teasers to avoid these linguistic traps, but that doesn't mean we don't still fall for them every once i…

That’s less a brain teaser than running into the error correction people use with language. This is useful when you simply can’t hear someone very well or when the speaker makes a mistake, but fails when language is intentionally misused.

  > This is useful when you simply can’t hear someone very well or when the speaker makes a mistake
I have a few friends with pretty heavy accents and broken English. Even my partner makes frequent mistakes as a non native English speaker. It's made me much better at communicating but it's also more work and easier for miscommunication to happen. I think a lot of people don't realize this also happens with variation in culture. So even within people speaking the same language. It's just that the accent serves as a flag for "pay closer attention". I suspect this is a subtle but contributing problem to miscommunication on the and why fights are so frequent.

Re: I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

#860

Post by bigfishrunning »

moogly wrote: post #711
viking123 wrote: post #117

Earlier quoted context omitted.

Lmao, and this is what they are saying will be an AGI in 6 months?

They seem to have stopped talking about AGI and pivoted to ads and smut.

This was probably wise, because ads and smut are well understood, and known to exist.
Post reply on HN