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

Vision language models are blind

vlmsareblind.github.io

Post reply on HN
Print view View original

111–120 of 202 posts

Re: Vision language models are blind

#111

Post by verdverm »

jampekka wrote: post #78
simonw wrote: post #62

Earlier quoted context omitted.

I see this complaint about LLMs all the time - that they're advertised as being infallible but fail the moment you give them a simple logic puzzle or ask for a citation. And yet... every interface to every LLM has a "ChatGPT can make mistakes. Check important info." style disclaimer. The hype around this stuff may be deafening, but it's often not entirely the direct fault of the model vendors themselves, who even put…

A bit like how Tesla Full Self-Driving is not to be used as self-driving. Or any other small print. Or ads in general. Lying by deliberately giving the wrong impression.

It would have to be called ChatAGI to be like TeslaFSD, where the company named it something it is most definitely not

Re: Vision language models are blind

#112

Post by fennecbutt »

cs702 wrote: post #13
tensor wrote: post #9

Earlier quoted context omitted.

I don't see how this is "embarrassing" in the slightest. These models are not human brains, and the fact that people equate them with human brains is an embarrassing failure of the humans more than anything about the models. It's entirely unsurprising that there are numerous cases that these models can't handle that are "obvious to humans." Machine learning has had this property since its invention and it's a classic…

It is embarrassingly, shockingly bad, because these models are advertised and sold as being capable of understanding images. Evidently, all these models still fall short.

Why do people expect these models, designed to be humanlike in their training, to be 100% perfect?

Humans fuck up all the time.

Re: Vision language models are blind

#113

Post by codeulike »

mkoubaa wrote: post #103

They interact with pixel buffers as a mathematical array. To call them blind is to confuse what they doing with the experience of sight...

Humans 'see' by tightly packed rods and cones in the retina sending signals up the optic nerve. Not as tidy as a mathematical array but nonetheless not all that different. Ultimately what comes to the brain from the retina can be thought of as a data structure of sorts.

Re: Vision language models are blind

#116

Post by jordan_bonecut »

This is an interesting article and goes along with how I understand how such models interpret input data. I'm not sure I would characterize the results as blurry vision, but maybe an inability to process what they see in a concrete manner.

All the LLMs and multi-modal models I've seen lack concrete reasoning. For instance, ask ChatGPT to perform 2 tasks, to summarize a chunk of text and to count how many words are in this chunk. ChatGPT will do a very good job summarizing the text and an awful job at counting the words. ChatGPT and all the transformer based models I've seen fail at similar concrete/mathematical reasoning tasks. This is the core problem of creating AGI and it generally seems like no one has made any progress towards synthesizing something with both a high and low level of intelligence.

My (unproven and probably incorrect) theory is that under the hood these networks lack information processing loops which make recursive tasks, like solving a math problem, very difficult.

Re: Vision language models are blind

#117

Post by jordan_bonecut »

mglz wrote: post #28

I tought some Computational Geometry courses and efficiently computing the intersections of N line segments is not as straightforward as you might initially think. Since somewhere some computation must be done to recognize this and LLMs are not specifically trained for this task, it's not suprising they struggle. In general, basic geometry seems under-explored by learning.

Yes, but so is telling if a photo contains a dog or understanding sentiment in a paragraph of text. Complexity isn't quite the issue, I think it is that there is a distinction between the type of reasoning which these models have learnt and that which is necessary for concrete mathematical reasoning.

Re: Vision language models are blind

#118

Post by aaron695 »

sweezyjeezy wrote: post #8

Entertaining, but I think the conclusion is way off. > their vision is, at best, like that of a person with myopia seeing fine details as blurry is a crazy thing to write in an abstract. Did they try to probe that hypothesis at all? I could (well actually I can't) share some examples from my job of GPT-4v doing some pretty difficult fine-grained visual tasks that invalidate this. Personally, I rate this paper [1], wh…

[dead]

Re: Vision language models are blind

#119

Post by scarface_74 »

jordan_bonecut wrote: post #116

This is an interesting article and goes along with how I understand how such models interpret input data. I'm not sure I would characterize the results as blurry vision, but maybe an inability to process what they see in a concrete manner. All the LLMs and multi-modal models I've seen lack concrete reasoning. For instance, ask ChatGPT to perform 2 tasks, to summarize a chunk of text and to count how many words are in…

Out of curiosity, I tried your test with ChatGPT 4o

https://chatgpt.com/share/79c5c6e1-e6a9-441b-acb3-54882303a8...

Of course as usual, LLMs are horrible with Math.

Funny enough, the next time it verified the word count by counting it out until I specifically told it to use Python

https://chatgpt.com/share/79e7b922-9b0f-4df9-98d0-2cd72d7041...

Re: Vision language models are blind

#120

Post by spookie »

youssefabdelm wrote: post #107
Lerc wrote: post #106

Earlier quoted context omitted.

I disagree. I think the title, abstract, and conclusion not only misrepresents the state of the models but it misrepresents Thier own findings. They have identified a class of problems that the models perform poorly at and have given a good description of the failure. They portray this as a representative example of the behaviour in general. This has not been shown and is probably not true. I don't think that models…

Exactly... I've found GPT-4o to be good at OCR for instance... doesn't seem "blind" to me.

You don't really need a LLM for OCR. Hell, I suppose they just run a python script in its VM and rephrase the output.

At least that's what I would do. Perhaps the script would be a "specialist model" in a sense.

Post reply on HN