yorwba

15 ms·

301.
by yorwba 5mo ago
MoE models are typically designed for datacenter deployment, where per-token load-balancing is more important, but it's also possible to use a different training objective that encourages domain-specialization of experts: https:/
302.
by yorwba 5mo ago
> I can't donate to Firefox or Thunderbird specifically, neither at the feature or product level. There is no way to ensure my donations go to enriching these products, instead of profit generating features that benefit the Mozilla
303.
by yorwba 5mo ago
The article includes that formula too and takes the overlap into account in its calculations.
304.
by yorwba 5mo ago
The Wikipedia article on Hangul has been rewritten over the years to deemphasize the evidence for a not-purely-de-novo origin, but you can still find it if you click the right links: https://en.wikipedia.org/wiki/Origin
305.
by yorwba 5mo ago
The intrinsic limitation of text diffusion is that natural text contains serial dependencies where a word at the beginning of the text strongly influences what comes later, and if there is a long enough dependency chain within a diffusion b
306.
by yorwba 5mo ago
Taking notes during a lecture is a neat trick to force the content to stay in short-term memory for at least a little while. But that also comes with the risk of transmission errors. Instructor-provided lecture notes can serve as a canonica
307.
by yorwba 5mo ago
Have you tried teaching without any textbook at all? Because that's how it worked during my CS education in Germany. All course content was written up in the lecture notes provided on the course homepage, variously a neatly-formatted L
308.
by yorwba 5mo ago
An LLM is not a human and in particular doesn't perform at a consistent level of ability like a human would. So past performance is only a weak indicator of future performance, even without changes to the underlying model. It was alway
309.
by yorwba 5mo ago
On the author's Flickr, the first image is in the "DASA Working World Exhibition" album: https://www.flickr.com/photos/mwichary/55316837619/in/album-... It's in Dortmund: https:&#x2F
310.
How to build a cancer vaccine, and whether they will work this time(owlposting.com)
2 points by yorwba 5mo ago|0 comments
311.
by yorwba 5mo ago
Those are somewhat separate concerns. You could have companies making independent hiring decisions while systematically discriminating against demographic groups, and you could also have companies all use the same system that systematically
312.
by yorwba 5mo ago
I didn't include it in my summary (it took me an hour to read the whole thing, obviously a lot had to be cut) but the article does actually address the "high resolution" argument in a three-paragraph bullet point under the &q
313.
by yorwba 5mo ago
A biological synapse's weight takes effect whenever its input changes. So although it cannot be copied and applied in parallel to different inputs at the same time (and hence your visual cortex has a bunch of more-or-less identical edg
314.
Efficient and Training-Free Single-Image Diffusion Models(arxiv.org)
52 points by yorwba 5mo ago|0 comments
315.
by yorwba 5mo ago
An attempt at a summary of the argument: - Human brains are estimated to have a few hundred trillion synapses. If you tried to replicate this in a neural network model with one parameter per synapse, it would be much larger than the largest
316.
by yorwba 5mo ago
The &emdash; is probably human error, other parts of the HTML correctly use — or Unicode em-dashes. Also: https://github.com/alexispurslane/rsync-analysis/commit/740b...
317.
by yorwba 5mo ago
I think you might be confusing the Rokkasho Reprocessing Plant (not yet operational, intended for plutonium extraction from spent fuel) and the Rokkasho Uranium Enrichment plant, which has been running at 75 tSWU/year (I think that sho
318.
by yorwba 5mo ago
Proposition 16. UHATs have polynomially bounded expansion over LTL. In particular, given an LTL formula φ, one can construct in polynomial time a UHAT T such that L(T) = L(φ). i.e. the blowup is only exponential in one direction.
319.
by yorwba 5mo ago
You can add stocks whenever you put money in. Whether that's because you got your paycheck or a dividend or some other income is kind of irrelevant. And you can remove stocks when you take money out. But you probably shouldn't sta
320.
by yorwba 5mo ago
If you pick stocks with the correct weight to track the index, you're effectively running an index fund. And so you don't have to rebalance to keep tracking the index.
321.
by yorwba 5mo ago
I thought the point of index funds weighting by market cap is that they don't require rebalancing, because the weight of stocks in the index exactly tracks price movements. You just keep holding the exact same number of shares, and mor
322.
by yorwba 5mo ago
> number array of x long (not sure how long x is, or if it's variable), which then gets projected down to a token representing that space (3-4 long as alpha-numeric) There is no such projection step. The array of x numbers is the
323.
by yorwba 5mo ago
This is now limited to only some restricted industries: https://www.ndrc.gov.cn/xxgk/zcfb/ghxwj/202504/P020250424307... Yes, the list is long, but it's a significant improvement to the before times
324.
by yorwba 5mo ago
It's possible to simulate the classical physics of fairly large game worlds using fairly small classical computer. If you wanted to model it using quantum physics instead (where quantum computers would theoretically have an advantage)
325.
by yorwba 5mo ago
The technical report https://microsoft.ai/wp-content/uploads/2026/06/main_2026060... has a lot of detail about decontaminating their training data and developing new in-house benchmarks to ensure reliabl
326.
by yorwba 5mo ago
There's no training code because the author is using an external service for that https://docs.primeintellect.ai/hosted-training/getting-start... The reward function is https://github.com/HarleyCoo
327.
by yorwba 5mo ago
If you give the model access to specialized tools (e.g. web search for question answering) the knowledge doesn't have to be stored in the model weights, which leaves some room for improvement. You'd still be overfitting to benchma
328.
by yorwba 5mo ago
The article explicitly addresses this, adjusts US heat deaths upward using a comparable method to the European figures, and still finds that there are many more heat deaths in Europe. E.g. from your first link: > ... some researchers say
329.
by yorwba 5mo ago
Europe not only has plenty of old people, they're also unlikely to have AC at home, so every year during a heat wave, a large number of them (though still a small share of the population) don't make it.
330.
by yorwba 5mo ago
Those threads have lots of comments that are both obviously LLM-generated and also flagged by users (e.g. https://news.ycombinator.com/item?id=48262928 ) So if anything, the spam filter is still not strict enough. And if a

More ›