I gotta say - processing video at 10fps is very impressive.
Gemini 3 Pro: the frontier of vision AI
231–240 of 309 posts
Re: Gemini 3 Pro: the frontier of vision AI
#232Earlier quoted context omitted.
Love how employee portals for many companies essentially never get updated design wise over the decades, lol. That page styling and the balls certainly take me back.
Literally decades: the login page looked like that when I joined google in 2007.
Except for the updated Google logo.
Re: Gemini 3 Pro: the frontier of vision AI
#233I'm playing with this and wondering if this is an actually good way to identify dominant colors and other features of a garment/product when using a photo where the item is styled and not isolated from the model or other garments
There should be an existing simpler way to do it. Image contains a bunch of pixels so you could just group the to see main colors
Re: Gemini 3 Pro: the frontier of vision AI
#234Earlier quoted context omitted.
"AI could never replace the creativity of a human" "Ok, I guess it could wipe out the economic demand for digital art, but it could never do all the autonomous tasks of a project manager" "Ok, I guess it could automate most of that away but there will always be a need for a human engineer to steer it and deal with the nuances of code" "Ok, well it could never automate blue collar work, how is it gonna wrench a pipe i…
I still haven't witnessed a serious attempt at passing the Turing test. Are we just assuming its been beaten, or have people tried? Like if you put someone in an online chat and ask them to identify if the person they're talking to is a bot or not, you're telling me your average joe honestly can't tell? A blog post or a random HN comment, sure, it can be hard to tell, but if you allow some back and forth.. i think we…
A couple of months ago I saw a paper (can't remember if published or just on arxiv) in which Turing's original 3-player Imitation Game was played with a human interrogator trying to discern which of a human responder and an LLM was the human. When the LLM was a recent ChatGPT version, the human interrogator guessed it to be the human over 70% of the time; when the LLM was weaker (I think Llama 2), the human interrogator guessed it to be the human something like 54% of the time.
IOW, LLMs pass the Turing test.
Re: Gemini 3 Pro: the frontier of vision AI
#235Earlier quoted context omitted.
I thought adversarial testing like this was a routine part of software engineering. He's checking to see how flexible it is. Maybe prompting would help, but it would be cool if it was more flexible.
So the idea is what? What's the successful outcome look like for this test, in your mind? What should good software do? Respond and say there are 5 legs? Or question what kind of dog this even is? Or get confused by a nonsensical picture that doesn't quite match the prompt in a confusing way? Should it understand the concept of a dog and be able to tell you that this isn't a real dog?
You know, I had a potential hire last week, and I was interviewing this one guy whose resume was really strong, it was exceptional in many ways plus his open-source code was looking really tight. But at the beginning of the interview, I always show the candidates the same silly code example with signed integer overflow undefined behavior baked in. I did the same here and asked him if he sees anything unusual with it, and he failed to detect it. We closed the round immediately and I disclosed no hire decision.
Re: Gemini 3 Pro: the frontier of vision AI
#236Earlier quoted context omitted.
Interpolation and generalization are two completely different constructs. Interpolation is when you have two data points and make a best guess where a hypothetical third point should fit between them. Generalization is when you have a distribution which describes a particular sample, and you apply it with some transformation (e.g. a margin of error, a confidence interval, p-value, etc.) to a population the sample is…
I assumed you were speaking by analogy, as LLMs do not work by interpolation, or anything resembling that. Diffusion models, maybe you can make that argument. But GPT-derived inference is fundamentally different. It works via model building and next token prediction, which is not interpolative. As for bias, I don’t see the distinction you are making. Biases in the training data produce biases in the weights. That’s w…
> It works via model building and next token prediction, which is not interpolative.
I'm not particularly well-versed in LLMs, but isn't there a step in there somewhere (latent space?) where you effectively interpolate in some high-dimensional space?
Re: Gemini 3 Pro: the frontier of vision AI
#237Earlier quoted context omitted.
An interesting test in this vein that I read about in a comment on here is generating a 13 hour clock—I tried just about every prompting trick and clever strategy I could come up with across many image models with no success. I think there's so much training data of 12 hour clocks that just clobbers the instructions entirely. It'll make a regular clock that skips from 11 to 13, or a regular clock with a plaque saying…
Right you are. It can do 26 hours just fine, but appears completely incapable when the layout would be too close to a normal clock. https://gemini.google.com/share/b3b68deaa6e6 I thought giving it a setting would help, but just skip that first response to see what I mean.
"just fine" is not really an accurate description of that 26-hour clock
Re: Gemini 3 Pro: the frontier of vision AI
#238Earlier quoted context omitted.
I assumed you were speaking by analogy, as LLMs do not work by interpolation, or anything resembling that. Diffusion models, maybe you can make that argument. But GPT-derived inference is fundamentally different. It works via model building and next token prediction, which is not interpolative. As for bias, I don’t see the distinction you are making. Biases in the training data produce biases in the weights. That’s w…
> It works via model building and next token prediction, which is not interpolative. I'm not particularly well-versed in LLMs, but isn't there a step in there somewhere (latent space?) where you effectively interpolate in some high-dimensional space?
Not interpolation, no. It is more like the N-gram autocomplete used to use to make typing and autocorrect suggestions in your phone. Attention js not N-gram, but you can kinda think of it as being a sparsely compressed N-gram where N=256k or whatever the context window size is. It’s not technically accurate, but it will get your intuition closer than thinking of it as interpolation.
The LLM uses attention and some other tricks (attention, it turns out, is not all you need) to build a probabilistic model of what the next token will be, which it then sampled. This is much more powerful than interpolation.
Re: Gemini 3 Pro: the frontier of vision AI
#239Earlier quoted context omitted.
I assumed you were speaking by analogy, as LLMs do not work by interpolation, or anything resembling that. Diffusion models, maybe you can make that argument. But GPT-derived inference is fundamentally different. It works via model building and next token prediction, which is not interpolative. As for bias, I don’t see the distinction you are making. Biases in the training data produce biases in the weights. That’s w…
What I meant was that what LLMs are doing is very similar to curve fitting, so I think it is not wrong to call it interpolation (curve fitting is a type of interpolation, but not all interpolation is curve fitting). As for bias, sampling bias is only one many types of biases. I mean the UNIX program YES(1) has a bias towards outputting the string y despite not sampling any data. You can very easily and deliberately p…
Training is kinda like curve fitting, but inference is not. The inference algorithm is random sampling from a next-token probability distribution.
It’s a subtle distinction, but I think an important one in this case, because if it was interpolation then genuine creativity would not be possible. But the attention mechanism results in model building in latent space, which then affects the next token distribution.
Re: Gemini 3 Pro: the frontier of vision AI
#240Going to compare this to our current solution of Amazon's Textract service for analyzing handwritten datasheets. Textract, when extracting tables (which is what we use it for) does not allow for providing any context or information about the tables and what we expect them to contain, but it is really good at correctly recognizing hand written characters. All of my attempts at less specialized, more general models all…
2.5 pro is already excellent at this.