Kye

27 ms·

511.
by Kye 2y ago
I wonder if this would be an ideal use case for diffusion-based LLMs slowly piecing it together with each pass from what it could determine from the last.
512.
by Kye 2y ago
I never used it because I could never figure out the pricing. Fortuitous.
513.
AI-assisted creativity and the race to claim the future(kyefox.com)
2 points by Kye 2y ago|0 comments
514.
by Kye 2y ago
Being able to quickly get a script for some simple automation, defining source and target formats in plain English, has been a huge help. There is simply no way I'm going to remember all that stuff as someone who doesn't program r
515.
by Kye 2y ago
It's the opposite incentive to ad-funded social media. One wants to drain your wallet and keep you hooked, the other wants you to spend as little of their funding as possible finding what you're looking for.
516.
by Kye 2y ago
>> "A fully loaded 128GB with a Ryzen AI Max+ 395 configuration (16 CPU cores, 40 GPU cores) will run you $1,999." That's a no-brainer to replace my laptop-as-desktop in the next few years. By then, I expect this will
517.
by Kye 2y ago
Yeah, I was going to say "through to what?" Enough people have to have enough money to be able to buy the things made in a factory someone's mom spent 30 years on for that factory to exist. These tools are being used to disma
518.
by Kye 2y ago
Applying those practices to western employment will mean no one can afford what those grateful people make.
519.
by Kye 2y ago
No, we know, and we're not okay with it either. I understand it's often an improvement over other options for employment. That makes it understandable and even supportable to some extent, not okay . Here in the US, we spent centu
520.
by Kye 2y ago
I thought o3-mini was o1-mini. OpenAI's naming gets confusing.
521.
by Kye 2y ago
Can you expand on that? I don't see where it could emerge from.
522.
by Kye 2y ago
A toddler can learn by trial and error mid-process. An LLM using autoregressive inference can only compound errors. The LLDM model paper was posted elsewhere, but: https://arxiv.org/pdf/2502.09992 It basically uses the
523.
by Kye 2y ago
> "Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize."
524.
by Kye 2y ago
It depends on how much you want it to do for you. I've used ChatGPT to come up with song briefs which I then turn into music myself.
525.
by Kye 2y ago
It seems like the torrent was already happening and DeepSeek's part is just one example of that. They did help bring attention to those advancements, and that's led to lots more people contributing and finding more niche applica
526.
by Kye 2y ago
The conclusion is basically what tech ethicists have said all along: it's all ethically troubled. Every bit. We spend each tech revolution singling out the latest bit while ignoring the conversations we had about the last bit, which we
527.
AI is an Ethical Nightmare (2023) [video](youtube.com)
2 points by Kye 2y ago|1 comments
528.
by Kye 2y ago
It does seem like some of the messaging gets away from their actual documentation: https://platform.openai.com/docs/guides/reasoning-best-pract... The reasoning models improve on this kind of thing, but that just
529.
by Kye 2y ago
It mostly just uses those other solutions in the background. You can open the Analyzing and see it building a script with OpenCV for vision tasks. It's a handy front-end.
530.
State of Create 2025 [pdf](stateofcreate.co)
1 points by Kye 2y ago|0 comments
531.
by Kye 2y ago
Sometimes it helps to have it outline the task first with a prompt like "Suggest good guidelines for [task]" and then "Follow those guidelines for [task]" once you check to make sure the guidelines are good. LLMs aren&#x
532.
by Kye 2y ago
I didn't say they nailed it. I don't know since I can't run it, but ten years is a long time for any technology. I toyed with the 7B Mistral through MLC Chat and, while slow, the responses were good. The Llama-3.2-3B-Instruct
533.
by Kye 2y ago
I could only skim to verify an automated summary, but if the takeaway is "these AI giants are doomed" then he's right. The future, probably within 10 years, is most tasks being handled by small on-portable-device models (7B p
534.
by Kye 2y ago
So are many erotica readers!
535.
by Kye 2y ago
Magic And fine tuning. Choose your fighter... High level overview: https://www.datacamp.com/tutorial/fine-tuning-large-language... More detail: https://www.turing.com/resources/finetuning-large-lan
536.
by Kye 2y ago
My standard prompts when I want thoroughness: "Did you miss anything?" "Can you fact check this?" "Does this accurately reflect the range of opinions on the subject?" Taking the output to another LLM with the s
537.
by Kye 2y ago
It varies a lot for me. One day it takes scattered documents, pasted in, and produces a flawless summary I can use to organize it all. The next, it barely manages a paragraph for detailed input. It does seem like Google is quick to respond
538.
by Kye 2y ago
The notion of a correct spelling is a relatively new innovation. I can't come up with a defense even for a conventional spelling that doesn't lean on "people will try to shame you for it," and that doesn't seem like
539.
by Kye 2y ago
I assumed your misread and mischaracterization wasn't with ill intent since I could have worded that better, ignored it aside from the clarifying edit, and focused on the bit that was interesting.
540.
by Kye 2y ago
edit: I mistook soapboxing for sincere interest in discussion. Please disregard.

More ›