qnleigh

19 ms·

271.
by qnleigh 1y ago
But were humans involved in picking which answer/shot to submit, or was it only AI?
272.
by qnleigh 1y ago
100 watts seems very low. A single Nvidia GeForce RTX 5090 is rated at ~600 watts. Probably they are using many GPUs/TPUs in parallel.
273.
by qnleigh 1y ago
This makes me wonder how much consumer spending as a whole is on products that people never use. It's not like this psychology is unique to gaming. Could it be as high as 50%?
274.
by qnleigh 1y ago
Literally thousands?? A 40-hour work week for a year is ~2000 hours. Out of curiosity, what's your rough 'playing schedule'?
275.
by qnleigh 1y ago
I think the point is that, on the whole, sales manipulate consumers into spending -more- money than they otherwise would have. Of course this isn't true for every person all the time, but on average I suspect it is; sales are common pa
276.
by qnleigh 1y ago
I'm still wrapping my head around the consequences of this, which almost seem too big to believe. It sounds like there are hundreds of millions if not billions to be made on filing these kinds of lawsuits, which means change is coming
277.
by qnleigh 1y ago
An observation: most articles with titles of the form "A SLAMS B" put forward a narrow, one-sided view of the issue they report on. Oftentimes they're shallow attempts to stir outrage for clicks. This one is just giving a CEO
278.
by qnleigh 1y ago
For VR goggles, I agree. But if we're headed toward AR glasses, most (all?) technologies in that space don't support opaque screen content. They project images into your eyes, but they can't block light coming in from the rea
279.
by qnleigh 1y ago
Yes that's still a great book, though it's starting to get a bit outdated. Some recent developments that would belong in an updated edition: - The section on error correction is still gold, but it doesn't cover "scalable
280.
by qnleigh 1y ago
Do we know how these apps were able to track browser activity? The only clues I see in the article are that it was on a per-website basis, and that it worked in incognito mode. I'm especially curious if Google shares any of the blame.
281.
by qnleigh 1y ago
This is a great summary. Is it plausible that the discrepancy between the two approaches points toward something else beyond the Standard Model? My impression from reading was 'probably not, because it would seem very ad-hoc and unusua
282.
by qnleigh 1y ago
I'm going to give him the benefit of the doubt and assume that he calls himself a survivor in response to accusations of having a 'victim mindset,' but not during SWE interviews...
283.
by qnleigh 1y ago
Ah yeah there are multiple definitions of 'acceleration' here. Unruh radiation occurs when you're not 'in an inertial reference frame,' loosely meaning that you feel acceleration. So in a rocket in space or (presuma
284.
by qnleigh 1y ago
Is there a simple way to understand why massive objects don't radiate gravitationally? Accelerating observers see a bath of thermal radiation via something called the Unruh effect. If you're standing on a planet, you're accel
285.
by qnleigh 1y ago
That's the first I've heard of this. Who thinks that?
286.
by qnleigh 1y ago
How well did VADER correlate with Llama? Did you try any other methods intermediate between them?
287.
by qnleigh 1y ago
Will there be an appeal? How finalized is this?
288.
by qnleigh 1y ago
I'm very happy to see that they're prioritizing making tools for musicians, rather than making AI music generators to replace musicians. Everything else I've come across so far was trying to do the latter.
289.
by qnleigh 2y ago
Do we know that this is how the algorithm actually works? The article only shows one plot of one specific instance, and there could be more than one explanation for the sudden drop in viewership (especially given the involvement of Twitter&
290.
by qnleigh 2y ago
There are lots of recorded lectures by him online too, which are great. His quantum field their lectures are especially popular https://www.damtp.cam.ac.uk/user/tong/qftvids.html .
291.
by qnleigh 2y ago
I'm confused; is this a proposed string-theoretic model of a black hole? Or is this a new object that they propose to form instead of a black hole when a star collapses?
292.
by qnleigh 2y ago
This is so on-point. Many things that we now take for granted from LLMs would have been considered sufficient evidence for AGI not all that long ago. Likely the only test of AGI is whether we can still come up with new goalpost.
293.
PromoPlot: Covering open-access fees by filling wasted space in corner plots(arxiv.org)
2 points by qnleigh 2y ago|1 comments
294.
by qnleigh 2y ago
Happy April 1st :)
295.
by qnleigh 2y ago
> use rockets to slowly launch the waste into space. We can already design a storage capsule that can survive re-entry. Do you have a source in mind for this claim? Even if a capsule could survive reentry, surely it wouldn't survive
296.
by qnleigh 2y ago
I think it's more that the Navy isn't constrained by Nimby's and anti-nuclear lobbyists.
297.
by qnleigh 2y ago
Well conservation of energy says you can't extract energy out of nowhere. In the paper they claim that the extracted energy comes at the price of slowing down Earth's rotation. But then this seems to violate conservation of angula
298.
by qnleigh 2y ago
This is a great question, and it's not as simple many comments here make it seem to be. This Veritasium video has a great explanation (and an enormous ball of molten sodium).
299.
by qnleigh 2y ago
That would make so much more sense. So then the comments about only slowing down Earth's rotation by a few ms/century. It would deplete Earth's magnetic field, and likely on a much faster timescale. This would not be good...
300.
by qnleigh 2y ago
Yes, but when you consider the combined system of the earth and everything that affects its orbit, angular momentum is conserved.

More ›