tensor

33 ms·

541.
by tensor 2y ago
Science works on consensus and iteration. Government policy generally (and I think should) follow that process, not form policies on minority opinions at the time. For instance, although the droplet vs airborne debate was incorrect initiall
542.
by tensor 2y ago
People disagreeing with you is not censorship.
543.
by tensor 2y ago
I'm going to need to see some evidence for this. How is it that I was reading science papers and media reports on them daily at the time, covering all these things you claim were censored? Droplet vs airborne was a frequent debate, as
544.
by tensor 2y ago
Some people's goal in life is to make money so that they can do nothing. Other people get pleasure from accomplishing things. I can't imagine doing something and not wanting to get better at it. What's the point? What's
545.
by tensor 2y ago
I thought the rule here was to match the article headline, which is "These fossils..".
546.
by tensor 2y ago
I see you mistyped Shotwell.
547.
by tensor 2y ago
To stop people dying. Which they did and continue to do.
548.
by tensor 2y ago
You are clearly trolling as this has been beaten to death. But for the sake of other readers, getting COVID has a far higher chance of clots than the vaccines. Also, the vaccines significantly reduced the chance of death, even if they ultim
549.
by tensor 2y ago
People frequently confuse freedom and consequences, but choices always have consequences in society. All schools have rules of conduct, if you violate them your kid gets kicked out. Jobs also have rules of conduct, dress code, etc. And yes,
550.
by tensor 2y ago
You already have the freedom to make your own decisions. What you don't have is the freedom to endanger others. That's when we make laws, and those laws may be informed by science but they are not created or enforced by scientists
551.
by tensor 2y ago
I'm not surprised at all. Honestly the "I don't need to learn that" mentality is common in tech even in people who call themselves senior developers. It's especially noticeable in the hostility of many towards the s
552.
by tensor 2y ago
The majority of people will not request your content. They’ll just consume the content that is easiest to access, and that content will be indexed using AI.
553.
by tensor 2y ago
Fully agreed on point A, but I've heard the "but then humans won't be trained" argument before and don't buy it. It's already the case that humans can cheat or get buy without fully understanding the math or id
554.
by tensor 2y ago
What? I have many many many monthly bills that are not iPhone apps. My gas bill may or may not have an iOS app, but I'm sure as hell not confused and angry if it doesn't.
555.
by tensor 2y ago
In the EDM world a ton of venues still only publish their schedule via insta.
556.
by tensor 2y ago
It's less a "large opinionated framework" and more "fewer components overall. There is no HTML/CSS for example, it goes direct from Dart (equivalent to JS) to the Widget rendering system." Dart has its own buil
557.
by tensor 2y ago
The Kagi AI agent does not. And a locally running one will not. Like it or not for those willing to pay the ads are going away.
558.
by tensor 2y ago
The new generation of AI search will likely use traditional indexes or RAG followed by the AI reading the results to do higher level filtering. If you look at how search gpt, kagi experts, or the hugging face llm search works I think that w
559.
by tensor 2y ago
Personally, I find it impossible to stomach continuing with ad-based models. I want zero ads in my life. Also, with newer AI the old search model is done for. Search will shift, should shift, to AI agents doing the search along with reading
560.
by tensor 2y ago
FWIW I was all in on Google years ago. But as features and products kept vanishing or degrading and being replaced with ad driven crap, it eventually drove me to swear off all Google products. The only one I still use is Workspace.
561.
by tensor 2y ago
It's important to note that outside world means the actual world, not the thoughts of other humans. You need a way to establish ground truth, which comes from observing the actual outcome of actions and experiments.
562.
by tensor 2y ago
If the AI generated content is filtered for quality or is corrected then it will still be good data. The phenomenon of model degradation is only in the case where there is no outside influence in the generated data.
563.
by tensor 2y ago
No, there is no vision LLM that produces confidence numbers to my knowledge.
564.
by tensor 2y ago
Traditional OCR neural networks like tesseract crucially they have strong measures of their accuracy levels, including when they employ dictionaries or the like to help with accuracy. LLMs, on the other hand, give you zero guarantees, and h
565.
by tensor 2y ago
No, disappearing = people thinking their data has more value than it has and setting their robots.txt so that it's no longer even indexed for search because they've banned any use of AI. Then, because it's actually not that v
566.
by tensor 2y ago
> It's about the copyright infringement to acquire the training set and it's about the copyright infringement when producing output. I just read your comment. I also read news articles I paid for today. None of that is copyrigh
567.
by tensor 2y ago
Art forgeries are an example of a human brain regurgitating the original content. Someone reading an article and rewriting it with some differences is a derivative work. So yes, a brain with knowledge is potentially infringing a copy or a d
568.
by tensor 2y ago
Synthetic data doesn't require using another ML models output as training data. One incredibly successful way of generating useful synthetic data is using rules. For example, AlphaZero uses synthetic data (playing against other version
569.
by tensor 2y ago
I don't see how this is "embarrassing" in the slightest. These models are not human brains, and the fact that people equate them with human brains is an embarrassing failure of the humans more than anything about the models.
570.
by tensor 2y ago
LaserCube[1] makes entry level lasers and has an open core [2]. There are also libraries that people have made that can interface directly with its wifi/ethernet protocol. However, programming lasers is not nearly as simple as doing 2d

More ›