disattention

I feel like this is my time to shine

I had the same thought, especially because of their recent collaboration.

https://pytorch.org/blog/pytorch-foundation-welcomes-ray-to-...

Seems inconsistent to say "truth be told" before "probably". These sorts of vibes based statements cloud conversation and confuse the scientifically illiterate who are quick to attach truth to belief.

In what way is food insecurity comparable to death by starvation?

There seems to be a clear difference between criminals doing shady things and the government doing shady things. It seems like a false equivalency to compare an incident where a random guy does something terrible to one where law enforcement is rapturing people into the night while wearing masks.

Anyone can literally just buy/make a uniform. Especially concerning if badges aren't visible. This argument doesn't hold much water imo.

This is a silly comment, and missing content that would make it helpful. At least say why you think these things are better than C. They want a language that allows them to explore the depths of systems, so they may very well want C or similar.

Please try to be more helpful and less presumptive, especially when someone is asking to learn.

The reason I replied to the parent comment was because of their dismissive tone. Of course returning the product is a good idea, but telling reasonable people to "chill out", and dismissing their concerns by suggesting they "just XYZ" is truly unproductive. It ends the conversation rather than engaging with it.

This is a kind of stoic virtue signal that may make people feel more mature for agreeing, but fails to fix issues while mocking people who try to make a difference. It's ok for people to feel things, and it's ok for people to want laws addressing anti-consumer behavior.

This exists in the US as well. I've personally experienced and witnessed it happen within labs at an R1 University. The accountability structures are woefully insufficient to protect students and junior researchers, and the incentives are perverse as to actually reinforce the practice.

I've seen frequently that talented technical contributors are academically handicapped because they bring too much value to the lab for them to graduate quickly. I've personally had my own funding threatened if I didn't work "at least 60 hours each week" on my ex-advisors work (which was in no way related to my degree or research interests). I was fortunate to find another advisor and funding source quickly, but most advisors are absolutely profiting in their career off the backs of their students; leveraging both carrot and stick to fuel their ambition. It's a problem of modern academia and I'm not sure how to fix it.

This is actually a good insight. It turns out that transformers are indeed a form of graph network, precisely because of the attention mechanism. Graph attention networks are actually a very popular GNN architecture. Generally, the issue with using an LLM style architecture for generic graphs is modeling the sparsity, but is possible by using the graph adjacency matrix to mask the attention matrix. There are a number of papers and articles which address this connection, and plenty of research into mechanisms for sparsifying attention in transformers.

There are also graph tokenizers for using more standard transformers on graphs for doing things like classification, generation, and community detection.

Yes? And it's not so much about the surveillance in the first place as opposed to the algorithmic manipulation of content to shift narratives and spread propaganda. Whether or not one believes the US based companies do similarly is beside the point, precisely because they're US based, whereas TikTok is the product of the US's primary economic and ideological adversary

It isn't unrealistic unfortunately. It's not unusual for a few people to rack up $300, especially with drinks involved. Many states (e.g. Texas) have sales tax, so already +8.25%. Add on an 18% automatic gratuity, then tip another 15% unknowingly and you're paying $440 for your $300 worth of food...

I think it's unwise to dismiss soft skills as getting in the way of that goal. As responsibility increases, and the complexity of your work increases, you'll have to lead people at some point. You can absolutely remain an überproductive IC forever, but you will cap out on what you can achieve -- and most importantly, what you can control -- by virtue of almost all great accomplishments being realized by more than one person.

Wanting to remain as technical as possible and be paid for your ability is valid, but soft skills will multiply your knowledge (and therefore value) across a team. In fact, I would argue soft skills are necessary to actualize your own potential as an IC within a team

This is a common pattern in zig and compliments comptime well. It's not supposed to be the constructor for the ArrayList, it's more like a C++ template in that you're constructing a type that then needs to be initialized. There isn't templating syntax in zig though, since you just write compile time code in zig directly (which is really nice).

Also, the type is different and ZLS will insert the hint if enabled.

Strangely pro-eugenics take...

Knowledge graphs have been around for a long time (early 70s). You could even argue they're older than that when considering something like an ontological map. I think this post is not meant to demonstrate novel technology or structure; rather, it's simply a self-reported introduction to the topic space.

The Wikipedia page just shows a linear time algorithm for identification of an undirected graph as an interval graph. The article notes that there's a polynomial time algorithm for coloring, which is the task-line assignment analogous problem. Still, not NP-hard, but not linear either unless there's another resource that mentions otherwise.

The license details are listed on the project GitHub

https://github.com/facebookresearch/seamless_communication#l...

Why not use existing OCR/document extraction tools [0]? There are a number of options, and even a custom implementation is probably a reasonable side project given some standardized structure.

[0] https://rossum.ai/lp/data-extraction