ianbicking

12 ms·

91.
by ianbicking 2y ago
They don't really need OpenAI. There's literally a dozen LLM providers that would be sufficient for what an LLM-powered Alexa would do. Amazon has their own LLMs too. The underlying LLM technology is hardly the issue, at least now
92.
by ianbicking 2y ago
Yeah, the basic premise is off because LLM responses are regularly tested against ground truth (like running the code they produce), and LLMs don't get to carefully select what requests they fulfill. To the contrary they fulfill reques
93.
by ianbicking 2y ago
This maybe points to another theory (which may be entirely wrong, I'm just guessing!): honeybees die because they aren't supposed to attack each other. Like they can't be aggressively selfish because they'll just die in
94.
by ianbicking 2y ago
It points to another evolutionary pressure that isn't mentioned as often: if an animal is too aggressive humans will exterminate it.
95.
by ianbicking 2y ago
I also thought it was interesting that it was actually several BASIC programs with data passed back and forth by stuffing it in specific memory locations.
96.
by ianbicking 2y ago
I notice if you look at Madrid it includes all of Spain and none of Portugal, and similarly from Lisbon. I assume this is because the schedules don't line up, because it wouldn't really make sense in terms of physical distance.
97.
by ianbicking 2y ago
"Langsmith appeared popular, but we had encountered challenges with Langchain from the same company, finding it overly complex for previous NonBioS tooling. We rewrote our systems to remove dependencies on Langchain and chose not to pr
98.
by ianbicking 2y ago
Here's a technical podcast interview with someone who is doing this: https://www.latent.space/p/nsfw-chatbots It starts feeling pretty weird when he starts talking about conversations where someone spends $1000.
99.
by ianbicking 2y ago
Oh! Format preserving editing too, very nice... https://github.com/hjson/hjson-js?tab=readme-ov-file#modify-...
100.
by ianbicking 2y ago
I like JSON5 and have used it some. When GPT was younger and I was parsing its JSON output directly, JSON5 was forgiving in useful ways. The one thing I really wish it had was some form of multi-line string, mostly so I could use it with li
101.
by ianbicking 2y ago
I'm not saying the ranking is necessarily wrong, but that turning the ranking into a distribution is constructed. And it MIGHT be a correct construction, but I am less confident that is true. The distribution implies something like &qu
102.
by ianbicking 2y ago
"IQ is Gaussian" – it was pointed out somewhere, and only then became obvious to me, that IQ is not Gaussian. The distribution is manufactured. If you have 1000 possible IQ questions, you can ask a bunch of people those questions,
103.
by ianbicking 2y ago
I have given Threads a good try, and recently when Bluesky activity started up I restarted using Bluesky (it didn't stick for me the first time). The technology doesn't really matter that much, as long as it's basically compe
104.
by ianbicking 2y ago
The repo doesn't say much... I thought maybe the docs would justify "world domination" in some fashion, but they are rather dry: https://haltp.org/posts/owl.html Is there something that describes what is
105.
by ianbicking 2y ago
I'm coming at this with no real JavaScript implementation knowledge, so these comments might not even make sense... The data sorting seems quite cleanly at first, but as I think more about it I don't quite get it. I guess you are
106.
by ianbicking 2y ago
Reminds me of the "Learn X the Hard Way" series, distributed as PDF I think, on the idea that if there's code samples you should transcribe them by hand because the act of transcribing matters. Maybe that's an argument f
107.
by ianbicking 2y ago
"when user expertise level is medium to high, and task complexity is low to medium" – this reminds me of Python itself. Python isn't the best at anything, it's slow, sometimes vague in its formalisms, etc. But it keeps b
108.
by ianbicking 2y ago
It would be interesting to think about how it got it wrong. My hunch is that in the "think step by step" section it made an early and incorrect conclusion (maybe even a subtly inferred conclusion) and LLMs are terrible at walking
109.
by ianbicking 2y ago
I think the best way to verify if a task could benefit from this is to try it by hand. What would a logical representation of legal requirements look like? If we can't model it by hand then we shouldn't expect an LLM to be able to
110.
by ianbicking 2y ago
I made a pipeline using Z3 (another prover language) to get LLMs to solve very specific puzzle problems: https://youtu.be/UjSf0rA1blc (and a presentation: https://youtu.be/TUAmfi8Ws1g ) Some thoughts: 1. Get
111.
by ianbicking 2y ago
A UI that might be relatively easy for the road problem: when you right click on an empty tile, in the left hand tile selector show all the tiles that would be "valid" in that position. You'd have to add metadata about how ti
112.
by ianbicking 2y ago
Bob Dylan got a Nobel prize in literature, which feels something like a precedence.
113.
by ianbicking 2y ago
The mink story is interesting... presumably they are native, but had disappeared locally due to overhunting. But probably many of their competitors also disappeared (martins, weasels, etc) so there's an opening. Though even that wouldn
114.
by ianbicking 2y ago
It's the return of the supercomputer! I really didn't think the supercomputer would come back as a thing, for so long it seemed stuck as a weird research project that only made sense for a tiny set of workloads... but it does make
115.
by ianbicking 2y ago
At 130 tons it seems like the only reasonable way to actually take possession is to transport it by rail (not self-powered, but presumably it's still mobile). But even then... it's still on a rail. Move it to a defunct spur where
116.
by ianbicking 2y ago
I've been hacking on a LLM-based text adventure this weekend: https://github.com/ianb/intra-game It's like the fourth or fifth time I've made an attempt at this, each time learning something along the wa
117.
by ianbicking 2y ago
Their Android integration certainly isn't on track to earn them any moats... https://hachyderm.io/@ianbicking/113099247306589777
118.
by ianbicking 2y ago
I'm in Minnesota, I'm just assuming California has the same problems. The issue is pretty nationalized. Some of it is students having a genuinely disturbing attachment to phones specifically, like a sense of anxiety when the phone
119.
by ianbicking 2y ago
In part I see this as the legislature stepping up to be the bad guy, and taking pressure off schools. School administrators know phones are a problem, but a minority of students and parents are very vocal and make it hard to put bans in pla
120.
by ianbicking 2y ago
Huh, that's a different use case than I was imagining. I actually don't know why I'd want a standard API from a frontend and backend that I control. In most applications where I make something chat-like (honestly a minority o

More ›