IanCal

33 ms·

661.
by IanCal 1y ago
Good lord. I think the change here will be something we've seen with the other modalities. Text was interestingly syntactically correct but nonsense sentences. Then paragraphs but the end of the article would go off the rails. Then the
662.
by IanCal 1y ago
They absolutely cannot be worse than the humans involved here. gpt-4o followed by a series of o models so you have gpt-4o and o4? Wonderful .
663.
by IanCal 1y ago
I've had better success with things like o3 with search, because it can actually go and read docs to help fix problems. It helped me dig through matrix specs, proposals and PRs and while the first suggestion didn't work (I thought
664.
by IanCal 1y ago
I agree on the UX. A few basic things seem totally broken. The flow of connecting a github account works, then disconnects, sometimes doesn't work, sometimes just errors. I can't install things that I could yesterday and my enviro
665.
by IanCal 1y ago
That's not what I've talked about them being for, but regardless it depends on the impact surely. If it can show you how someone may misunderstand your point and either help correct it or just show the problem then yes that can ea
666.
by IanCal 1y ago
There are so many tasks in the world that 1. Involve a computer 2. Do not require incredible intelligence 3. Involve the messiness of the real world enough that you can't write exact code to do it without it being insanely fragile LLMs
667.
by IanCal 1y ago
You can use a GPU with docker - at least on some platforms. There's more setup though, nvidia have some details to help https://docs.nvidia.com/datacenter/cloud-native/container-to...
668.
by IanCal 1y ago
Laws obviously vary by location but in the EU there is this: https://commission.europa.eu/law/law-topic/data-protection/r... Beyond that, it looks like this system gives you full audit logs and lets you backt
669.
by IanCal 1y ago
You seem to be mixing up using ai to make decisions and using ai to make the decision tree in code to a specification. I think you should take a moment to try and understand the product before attacking so strongly.
670.
by IanCal 1y ago
There are fundamentally two different things going on here. One is whether kids should get fed. The other is how this is paid for. The approach of "the child is charged like anyone else buying a thing and hopefully their parents have g
671.
by IanCal 1y ago
Rather reminds me of Pratchetts character Lu-Tze, who having seen so many travel to the monasteries to achieve enlightenment decides to travel to Ankh Morpork and learns many ancient wisdoms ('Is it not written "Oo, you are so sha
672.
by IanCal 1y ago
I tried the image without search and it talked about Dornbirn anyway but ended up choosing Bezau which is really quite close. edit - the models are also at a disadvantage in a way too, they don't have a map to look at while the pick th
673.
by IanCal 1y ago
This can be as simple as giving them search over communication with a PO, and giving them a place to store information that's searchable. How good they are at this is a different matter but the article claims it is impossible because
674.
by IanCal 1y ago
> The system you describe appears to treat programmers as mere cogs Not at all, it simply addresses key issues raised. That they cannot have a theory of the program because they are reading it and not actually writing it - so have them w
675.
by IanCal 1y ago
Skipping that they say it's fallacious at the start, none of the arguments in the article are valid if you simply have models 1. Run code 2. Communicate with POs 3. Iteratively write code
676.
by IanCal 1y ago
What's the purpose of this? > In this essay, I will perform the logical fallacy of argument from authority (wikipedia.org) to attack the notion that large language model (LLM)-based generative "AI" systems are capable of d
677.
by IanCal 1y ago
Told that it was a mate in 2 puzzle, and it solved it for me https://chatgpt.com/share/680f4a02-4cc4-8002-8301-59214fca78... It worked through some stuff then decided to try and list all possible moves as there can&#x2
678.
by IanCal 1y ago
I don't know, I didn't spot the answer and it's from a list of hard puzzles from a chess coach. The model also wasn't told it was mate in 2 (or even if a mate was possible), just to solve it and it was white to move. ht
679.
by IanCal 1y ago
> That seems distinctly unimpressive I cannot understate how impressive this is to me, having been involved in ai research projects and robotics in years gone by. This is a general purpose model, given an image and human written request
680.
by IanCal 1y ago
That's a fantastic idea! I love the little me on that - taken from just a webcam pic with a messy background and it's pulled out things like my tshirt perfectly.
681.
by IanCal 1y ago
I'm not arguing this is the purpose here but data augmentation has been done for ages. It just kind of sucks a lot of the time. You take your images and crop, shift, etc them so that your model doesn't learn "all x are in the
682.
by IanCal 1y ago
That's in the list. > Battery endurance in cycles.
683.
by IanCal 1y ago
But it's nothing like GDP.
684.
by IanCal 1y ago
Although those can be more complicated, and it should be clear what they're for and why they exist. Will this result in an object having an updated_at timestamp elsewhere in a larger application? Is it clear which properties that ref
685.
by IanCal 1y ago
60,000 seconds was the amount of time it was tested for, they then extrapolate that out linearly. It doesn't need refreshing that often.
686.
by IanCal 1y ago
What does chrome have to do with this? The headline is about chrome, the article text is about search and owning chrome wouldn't give you search.
687.
by IanCal 1y ago
I've always thought it should work the other way around too, people show diagrams where the lines get squished at the top then tell me the pressure had gone down which isn't intuitive.
688.
by IanCal 2y ago
You could run an llm deterministically too. We're often explicitly adding in randomness to the results so it feels weird to then accuse them of not being intelligent after we deliberately force them off the path.
689.
by IanCal 2y ago
As in aws is a much bigger security concern?
690.
by IanCal 2y ago
Good find. The DOI resolves and the article is up now too https://iopscience.iop.org/article/10.3847/2041-8213/adc1c8

More ›