c7b

33 ms·

271.
by c7b 4y ago
GPT?
272.
by c7b 4y ago
True, but 'self-replication, and self-improvement' don't seem that implausible. If future corpora include LLM training repos and/or weights, we're not too far from imagining that an LLM with some overarching archi
273.
by c7b 4y ago
One remark: .io looks like it's #3 in the time series, but it seems to be missing from bar plot.
274.
by c7b 4y ago
obligatory xkcd: https://xkcd.com/331/
275.
by c7b 4y ago
The study linked in the article says that 99% of MLM participants end up losing money. Unless you have some data to back up similar stats for startup employees, you're creating a false equivalence (and a startup failing isn't the
276.
by c7b 4y ago
Not sure I get your question right, but this points right to the HN comment page under <link> https://hnrss.org/frontpage?link=comments
277.
by c7b 4y ago
Oh wow, you built this. Congratulations, really impressive! Do you have a post somewhere about how you built those AIs?
278.
by c7b 4y ago
I believe David Guetta that he was really just playing around with these cool tools here (who isn't blown away by them?). But I guess it won't go unnoticed by the teams and investors of uberduck and other AI startups that this is
279.
by c7b 4y ago
In theory, there exist firm plans to do an orderly resolution that can result in a market exit should the bank not be able to survive: https://www.finma.ch/en/enforcement/recovery-and-resolution/... In practi
280.
AI will degrade the consumer experience(project-syndicate.org)
3 points by c7b 4y ago|0 comments
281.
by c7b 4y ago
It is a bit surprising that Google got scooped by OpenAI, but the product surely has existed before ChatGPT went public. LaMDA was the model behind the 'Our AI has become sentient' controversy last year, if you remember that [0].
282.
by c7b 4y ago
Oh wow, cool work! Yeah, the 'why' question can be hard to answer. We already have a proof that spells out a perfectly rigorous answer, but often we're looking for a 'deeper' connection to something perceived as mor
283.
by c7b 4y ago
Seems like this AI model doesn't understand very well how it works[0], which I suppose can be explained (maybe there aren't that many explanations in the corpus), but it's also quite ironic. [0] The answers focus on the techn
284.
by c7b 4y ago
Not too familiar with this specific work, but it's not so uncommon in math that proofs are first found using some very large inputs that are then progressively simplified in further work. One prominent example would be prime gaps: in 2
285.
by c7b 4y ago
The first question that I have when I see a tools like this is a different one, namely what about (inevitable) misclassifications? False accusations by an ML model, in a world where people are increasingly willing to trust such models, coul
286.
by c7b 4y ago
> There are to my recollection numerous stories of people being locked out of accounts which they have the passwords for because Google has decided that things are suspicious and having the password is not enough. As someone who used to
287.
by c7b 4y ago
> Imagine a world where the only content you see is from publishers that you trust, and that your friends trust, and their friends, to maybe 4 or 5 hops or so, and the feed was weighted by how much they are trusted by your particular so
288.
by c7b 4y ago
> AI generated content that gets posted to the web as original human generated content, with LLMs getting re-trained on this content Isn't that extrapolating the current trend a bit too much? Clearly, the text corpora[0] amassed bef
289.
by c7b 4y ago
I scrolled through the article looking for examples/instructions how to plot maps for places other than the US, but sadly that doesn't seem to be covered.
290.
by c7b 4y ago
I'm not a legal expert, but my impression is that pretty much all of his actions since this blew up have mainly hurt his case, probably badly. He even lives with his two Stanford lawyer parents now, do they not tell him or does he not
291.
by c7b 4y ago
"Incentivize to pass" the institution that makes and grades the tests - what effect do you think that will have?
292.
by c7b 4y ago
One thing that makes me really sad about the new design is that the links to the same article in different languages are now collapsed by default. Even when you don't want to read the article in another language (which I highly recomme
293.
by c7b 4y ago
Congrats, looks really amazing! May I ask what your AI stack is? Do you use MidJourney for the images and GPT for the stories or something else? How do you ensure the output stays consistent across images - do you use the seed image and a c
294.
by c7b 4y ago
I find that Midjourney tends to have its own characteristic touch as well. It's great if you like it, but it can be tricky if you want something different. I've found StableDiffusion to be a bit more versatile, even though it&#x27
295.
by c7b 4y ago
These look amazing! Did you build MagicStory?
296.
by c7b 4y ago
Great job! Did you use any special techniques for getting the same characters in different pictures/settings?
297.
by c7b 4y ago
A huge left arrow with a tiny right arrow inside of it - can't make that up.
298.
by c7b 4y ago
How do you do the sticker printing? I recently helped make some stickers for marketing for a friend's business, and I was surprised how much they cost to order in small batches (which is annoying, because we wanted to make miniseries w
299.
by c7b 4y ago
I think that's it, thank you!
300.
by c7b 4y ago
HN Title: Signal app's president-"layoffs in tech are to keep salaries in check" That's not what the linked tweet says. Not literally and not implicitly. The tweet literally says "It's not about $" - whic

More ›