sega_sai

13 ms·

211.
by sega_sai 1y ago
Are you saying that the factual content of the article is incorrect? If not, yes the author may have views on Musk, but that is irrelevant in the context of the article.
212.
by sega_sai 1y ago
Non-negative integer solutions
213.
by sega_sai 1y ago
After months of using a hammer, I'm back to using my hands. LLM is a tool. I think we still need to learn how to use it, and in different areas it can do either more or less stuff for you. Personally I don't it for most of everyd
214.
by sega_sai 1y ago
This is obviously completely bizarre approach for random internet people to pick up titles and judge whether it appropriate topic or not. First, a majority of the grants listed there are in the EDU Directorate, i.e. devoted to EDU Programs
215.
by sega_sai 1y ago
I've seen the book, but I've personally found it not very useful for a person who wants to first get the basics. The natural number's game is actually quite fun, and I did understand much better the language. And it's al
216.
by sega_sai 1y ago
Interesting. I always wanted to try Lean, and personally never found an easy way to do it, as it requires installing a plugin in vscode, create a project or reading the lean book. But following the links I've found this nice interacti
217.
by sega_sai 1y ago
I am not sure what you mean. The idea is that the network should return the text, and a confidence expressed as probability. When trained, the log-score should be optimized. (i'm not sure it would actually work given how the training i
218.
by sega_sai 1y ago
Can we have models also return a probability, reflecting how accurate the statements it made is ?
219.
by sega_sai 1y ago
I think there is a difference between not actually looking at the event/object because you want to make photos (or make a perfect photo), vs just ensuring that essential stuff you see got at least a photo. Nowadays I make a conscious e
220.
by sega_sai 1y ago
I think the unbiasedness is completely red herring here, but do I agree with the point on focusing on the tests separately and implementations separately. Ideally you'd want two completely different LLMs work on both. But I think the
221.
by sega_sai 1y ago
I am sure the attempts of hiring for government jobs will go well, after all the indiscriminate firings experienced by many government agencies thanks to the DOGE and company.
222.
by sega_sai 1y ago
They also do month-long astrophysics summer schools for students every two years or so. I was lucky to be part of the school ~20 years ago. We had classes in the papal residence in Castel Gandolfo and audience with John Paul II. It was a gr
223.
by sega_sai 1y ago
I think this is a great point! I.e. while for humans, it's easier to write not strongly-typed python-like code, as you skip a lot of boiler-plate code, but for AI, the boiler-plate is probably useful, because it reinforces what variab
224.
by sega_sai 1y ago
Currently we don't even know if type Ia is merger of two WDs, or accretion from a normal star to a WD. Also the light-curve of the Ia SN is determined by the specific decays of various elements (Nickel and so forth). The ratio of those
225.
by sega_sai 1y ago
For every sort of measurement, or constraint derived from data, you have uncertainties related to random noise (either related to do measurement error, or sampling error caused by finite number of objects in your sample). But you also have
226.
by sega_sai 1y ago
While it is possible that relaxing the homogeneity assumption is needed to resolve the tensions observed among different cosmological measurements, it is also worth noting that supernovae are a quite messy tracer, because we don't real
227.
by sega_sai 1y ago
Absolutely the same complaint. I wanted to see if it could summarize papers well, but I just could not handle all the conversation and attempts to make it 'exciting'. Especially in areas where I already know the background.
228.
by sega_sai 1y ago
It is one thing that you are getting results that are samples from the distribution ( and you can always set the temperature to zero and get there mode of the distribution), but completely another when the distribution changes from day to d
229.
by sega_sai 1y ago
With all the different open-weight models appearing, is there some way of figuring out what model would work with sensible speed (> X tok/s) on a standard desktop GPU ? I.e. I have Quadro RTX 4000 with 8G vram and seeing all the m
230.
by sega_sai 1y ago
From the article: "None of the changes, including falsified information about food allergens that could have been harmful to visitors, ever appeared before the public, according to court records."
231.
by sega_sai 1y ago
The paper certainly contradicts my expectation from the title. I.e. it does not present an LLM that can generate images without any access to images before.
232.
by sega_sai 2y ago
So he had a chance to say what he thinks of these plans to butcher NSF, but he chose to wash his hands and say "I am deeply grateful to the Presidents for the opportunity to serve our nation. " and so on.
233.
by sega_sai 2y ago
As a scientist, I actually greatly enjoy the AI assisted coding because it can help with the boring/tedious side of coding. I.e. I occasionally have some new ideas/algorithms to try, and previously I did not have enough time to
234.
by sega_sai 2y ago
It is annoying that while the connector itself is now finally standardized, the details of implementation diverge. So while you may be able to charge, it'll be slow charge.
235.
by sega_sai 2y ago
the chance of that is negligible.
236.
by sega_sai 2y ago
The size of the black hole described in the paper is ~ 20 km, so it is tiny. Even we have millions of such objects (and most likely we do), the chance of hitting something, given the enormous size of the galaxy is negligible.
237.
by sega_sai 2y ago
What do you expect when ICE have quotas on arrests -- https://www.newyorker.com/news/the-lede/the-mystery-of-ices-...
238.
by sega_sai 2y ago
I think many people are worried, but unless there is a good alternative theory, the particle is still the simplest explanation. Over last 20 years a lot focus was on WIMPs (weakly interacting massive particles), but those searches ruled out
239.
by sega_sai 2y ago
The only hope here, is that it will not happen. But cutting a mission like Nancy Grace telescope that has been in the works for > 10 years, 1 year before launch and on budget is peak idiocy. I know people who spend years on this mission,
240.
White House proposal would slash NASA science budget and cancel major missions(spacenews.com)
25 points by sega_sai 2y ago|2 comments

More ›