HackerTrans
TopNewTrendsCommentsPastAskShowJobs

theemathas

no profile record

Submissions

Generating Human-Level Text with Contrastive Search in Transformers

huggingface.co
1 points·by theemathas·vor 4 Jahren·0 comments

Stable Diffusion legal issues: Non-free license? Is it copyrightable?

youtube.com
2 points·by theemathas·vor 4 Jahren·0 comments

List of Stable Diffusion resources

rentry.co
164 points·by theemathas·vor 4 Jahren·23 comments

Character.ai (chatbot) drama due to censorship

twitter.com
9 points·by theemathas·vor 4 Jahren·3 comments

comments

theemathas
·vor 3 Jahren·discuss
Pyinstaller doesn't work with python 3.12.0. You need to use python 3.12.1. https://github.com/pyinstaller/pyinstaller/issues/7992
theemathas
·vor 3 Jahren·discuss
Sorry for the irrelevant reply, but I'm trying to get some attention on this.

This youtuber got a fraudulent DMCA notice, and his counter-notice got automatically rejected. The person who sent the notice isn't associated with the devs, and is asking for a $100 ransom to get the channel reinstated. How could this be fixed?

Proof that the person sending the notice isn't "authorized on behalf of the copyright holder": https://twitter.com/Lionheart/status/1671349063931838464

Tweet about the automatic rejection of the counter-notice: https://twitter.com/Azrial_Vanity/status/1671696220190810113

Tweet about the initial DMCA takedown: https://twitter.com/Azrial_Vanity/status/1671323777978736643

Mention of the $100 ransom: https://twitter.com/Azrial_Vanity/status/1671332970836140032

Any youtube people on here can help? Or anyone else have any advice?
theemathas
·vor 3 Jahren·discuss
This youtuber got a fraudulent DMCA notice, and his counter-notice got automatically rejected. The person who sent the notice isn't associated with the devs, and is asking for a $100 ransom to get the channel reinstated. How could this be fixed?

Proof that the person sending the notice isn't "authorized on behalf of the copyright holder": https://twitter.com/Lionheart/status/1671349063931838464

Tweet about the automatic rejection of the counter-notice: https://twitter.com/Azrial_Vanity/status/1671696220190810113

Tweet about the initial DMCA takedown: https://twitter.com/Azrial_Vanity/status/1671323777978736643

Mention of the $100 ransom: https://twitter.com/Azrial_Vanity/status/1671332970836140032

Any youtube people on here can help?
theemathas
·vor 3 Jahren·discuss
RLHF makes it possible to get good results with a bad prompt.
theemathas
·vor 3 Jahren·discuss
Probably stuff like this (note: possibly misleading URL) https://www.theverge.com/2023/2/15/23599072/microsoft-ai-bin...

Early Bing AI had a tendency to be rather unhinged.
theemathas
·vor 3 Jahren·discuss
A similar issue happened to Steam in 2015 https://www.forbes.com/sites/insertcoin/2015/12/25/steam-is-...
theemathas
·vor 3 Jahren·discuss
Reversing time on an attracting force still gives you an attractive force. Velocity is reversed, but acceleration isn't.

Imagine a ball being thrown up and then falling down, in a parabola. Reversing a video of that still gives you a video of a ball in a normal parabola trajectory.
theemathas
·vor 3 Jahren·discuss
> All real numbers, and all numbers of any other variety, can be written with a finite number of symbols.

This is false. In some sense, there exist numbers that can't be referred to. We can refer to the set of real numbers as a whole, but not some of the elements. https://en.wikipedia.org/wiki/Definable_real_number

But then there are issues with defining "definable numbers", which complicates things by a lot. https://mathoverflow.net/questions/44102/is-the-analysis-as-...
theemathas
·vor 3 Jahren·discuss
An alternative formulation of the axiom of choice: The cartesian product of a collection of non-empty sets is non-empty.
theemathas
·vor 3 Jahren·discuss
The main reason that we use real numbers is because we need real numbers to do calculus, at least in the normal way. And to do physics, we need to do calculus. You can keep approximating integrals as sums of very large number of terms, but it gets unwieldy to handle symbolically.
theemathas
·vor 4 Jahren·discuss
Seems really close to reality https://prompt-to-prompt.github.io/
theemathas
·vor 4 Jahren·discuss
You can give a blank prompt to stable diffusion and it will give you reasonable images https://www.reddit.com/r/StableDiffusion/comments/xu0dt6/no_...
theemathas
·vor 4 Jahren·discuss
Proper hands are already here https://www.reddit.com/r/StableDiffusion/comments/z8gje0/har...
theemathas
·vor 4 Jahren·discuss
Previous discussion on HN about the website: https://news.ycombinator.com/item?id=33020694

For the last few days, there seem to be a lot of new users discovering the website, due to word of it spreading within the vtuber community, and a Washington Post article.

Currently, the website has been down for almost an entire day now. I assume the devs are trying to figure out how to deal with this.

The devs apparently wrote a statement which then got deleted from some reason (TLDR: They say some of the issues were unintended side effects of changes made to handle the increase in users, and the devs are trying to fix the issues. There's also a separate set of issues that were side effects from a change intended to censor NSFW stuff.) https://www.reddit.com/r/CharacterAI/comments/y03ncu/newsupd...

Some users are extremely angry, and their discord got so bad that they had to (justifiably) shut down it completely (hopefully temporarily). And I've heard of harassment and even suicide threats (Can't find the source though).
theemathas
·vor 4 Jahren·discuss
Here's a similar case of a scanner using a traditional compression algorithm. It has a bug in the compression algorithm, which made it replace a number in the scanned image with a different number.

https://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres...
theemathas
·vor 4 Jahren·discuss
The o(1) here represents a decreasing function in m, such as 1/m. (Note: This is little o, not big o.)

Therefore, for any k > 1: m^(1+o(1)) grows more slowly than m^k but faster than m
theemathas
·vor 5 Jahren·discuss
Systems which are turing-complete but not turing-equivalent are stronger than turing machines. See: https://en.wikipedia.org/wiki/Hypercomputation

The "All known Turing-complete systems are Turing-equivalent, which adds support to the Church–Turing thesis." thing seems to have an implied "Turing-complete systems that are physically implementable." (I edited the article to add that.)
theemathas
·vor 6 Jahren·discuss
The video in the link says that an automatically adjusted hoop has better results than a manually adjusted hoop.