visarga

46 ms·

751.
by visarga 1y ago
> Conversations are error prone and noisy. I thought you'd say not being able to reload the form at a later time from the same URL is bad. This would be a "quantum UI" slightly different every time you load it.
752.
by visarga 1y ago
> it makes you not really capable of escaping the accept, accept, accept trap The definition of vibe coding - trust the process, let it make errors and recover
753.
by visarga 1y ago
You should try Cursor or Windsurf, with Claude or Gemini model. Create a documentation file first. Generate tests for everything. The more the better. Then let it cycle 100 times until tests pass. Normal programming is like walking, deliber
754.
by visarga 1y ago
> Imagine, for a moment, a world with no humans. Just machines, bolts and screws, zeros and ones. There is no emotion. There is no art. There is only logic. Humans use logic-defying algorithms called “emotions”. They get angry. They get
755.
by visarga 1y ago
Good method to generate synthetic training data, but only works for domains where validation can be scaled up.
756.
by visarga 1y ago
> I feel like I should understand the purpose of knowledge graphs, but I just... don't. Like RAG, it decouples KG size from context size, but unlike RAG, a KG offers deduplication and relational traversal. Some searches based on jus
757.
by visarga 1y ago
I find letting the agent iterate search leads to better results. It can direct the search dynamically.
758.
by visarga 1y ago
From this point of view XML offers all that and named brackets.
759.
by visarga 1y ago
Google benefits at scale from infrastructure, systems and laws that create the opportunity to make such revenues. It is only natural they should not harm the host.
760.
by visarga 1y ago
Art was not based on copyright for a long time, it won't disappear, just return to the natural way of sharing it. This trend started decades ago with internet, not with AI. A second fallacy is that AI art is public. Not true, most of i
761.
by visarga 1y ago
It worked allright for me when I was using it, a month ago. I cancelled because they somhow lost my paid credits and refused to refund me. No matter how great an AI tool, if the company is mismanaging user payments or usage tracking, it is
762.
by visarga 1y ago
I can't get Suno to obey the style tags, it's as if they are not used at all in generating the output, or massively changed. Maybe they do the same trick Gemini used to generate the black queen of England. Or it's a trick to
763.
by visarga 1y ago
You can use ChatGPT for these kinds of questions but it needs to use search or research mode, don't ask it in closed book mode.
764.
by visarga 1y ago
The homepage of YT has become a disaster area, it has almost zero customizability or exploration value. Can't remove vertical video shorts, can't control the topics displayed at the top. It maximizes engagement and time waste, not
765.
by visarga 1y ago
I think they are different systems, one is a collection of saved snippets and the other more like RAG over chat history.
766.
by visarga 1y ago
Let me show you the etymology of Bamba: SSM (state space model) -> SSSM (structured state space model) -> (it's like a snake ssss...) Mamba -> Bamba
767.
by visarga 2y ago
This is a very interesting approach. I naturally think this relates to LLMs, we could in principle encode a Differential DFA chain as additional inputs for AI models, so they could gain capabilities to run symbolic operations reliably, a Ne
768.
by visarga 2y ago
Yeah, tried it for a couple of minutes, $0.31, quickly stopped and moved away.
769.
by visarga 2y ago
> "They’re not really about coffee anymore. They’re selling a lifestyle." I like the emphatic style of GPT-4o lately. I can spot it from a mile.
770.
by visarga 2y ago
The situation with LLMs is much different than search, Google doesn't have such a large lead here. LLMs are social things, they learn from each other, any provider with SOTA model will see its abilities leaked through synthetic trainin
771.
by visarga 2y ago
In my professional opinion they can extract data at 85-95% accuracy.
772.
by visarga 2y ago
> Training LLMs on detailed interaction data from AI-powered IDEs could become a powerful flywheel leading to the automation of practical coding. I agree. But this is a more general flywheel effect. OpenAI has 500M users generating trill
773.
by visarga 2y ago
The idea is to use the LLM as a word smith, it doesn't conceptualize novel ideas from itself, that is your job. I would normally chat the topic until it gets it, and after that I ask it to write it up as an article. Here is a feed I co
774.
by visarga 2y ago
I think you should try to get the intent instead of stumbling on surface level. The core idea is that recursion explains emergence. A distributed system can still achieve centralized outcomes as a result of centralizing constraints acting o
775.
by visarga 2y ago
The irony of an article decrying AI influence, written with AI - but I feel where it is coming from. I used to watch long YT videos (lectures, debates) and read long articles, now I am more likely to get the transcript into a LLM and chat i
776.
by visarga 2y ago
Nothing happened. Judge by the outputs, if they are infringing, the model learned expression. If not, it learned abstractions and styles. Why do you think there were only 2-3 lawsuits focusing on output infringement? If regurgitation was a
777.
by visarga 2y ago
I am more fearful that Ghilibi will demand to own their style, forbidding anyone from using it without permission. If everyone stakes a claim on some style or abstraction, then we are in a situation where nothing new can be created without
778.
by visarga 2y ago
The idea is to let humans be humans, make a mess, debate, have their opinions, and AI comes after that and removes the herp derp from the useful parts. As a test of concept copy paste this whole page, put it in a LLM and ask for an article.
779.
by visarga 2y ago
I understand what you are saying, but there are different kinds of learning. If your model learns how many fingers are on a hand, it does not violate copyright in that regard. Every copyrighted work relies on facts, abstractions and styles
780.
by visarga 2y ago
It depends, is it a LoRa focused on Benn? Is it intended to replicate his style? Or is Benn one of the billion authors in the training set? In that case his impact will be minimal, and the generated images/texts will be very different

More ›