derbOac

7 ms·

61.
by derbOac 4mo ago
You might be completely correct, although my hunch is this is something that would require a change in architecture rather than increases in scale. The failure points happen in a fairly simple task (Stroop) with increases in repetition of t
62.
Deficient executive control in transformer attention(academic.oup.com)
40 points by derbOac 4mo ago|12 comments
63.
by derbOac 4mo ago
I'd love to see more of this sort of thing, as I think it's central to a lot of controversies in AI at the moment (not all but many).
64.
by derbOac 4mo ago
I don't think I'm smarter than anyone here but I think I've read about this before; I don't think it's just one study. There's also an emerging literature suggesting that deficits in specific amino acids are a
65.
by derbOac 4mo ago
There are lots of problems with this, but taking off UU is sort of ironic and shortsighted in particular, given that the UU community is probably a good source of chaplains given the nature of the religion. You can see this in the comments.
66.
by derbOac 4mo ago
Palantir is taking notes.
67.
Einstein's "wormhole" may reveal a hidden mirror of time(sciencedaily.com)
3 points by derbOac 5mo ago|0 comments
68.
by derbOac 5mo ago
This was portrayed in the film Contact (1997) closely to what's described in this piece (maybe it's in the book too? I haven't read it).
69.
Bipedalism and brain expansion explain human handedness(journals.plos.org)
1 points by derbOac 5mo ago|0 comments
70.
by derbOac 5mo ago
There's a peer-reviewed study of this that appeared in the last couple of years that showed the percent support for authoritarian rule tends to hover around 20% worldwide regardless of country, plus or minus some fuzz amount. I can&#x2
71.
by derbOac 5mo ago
That's the biggest risk. I can also imagine it encouraging a certain kind of fraud. The program sounds reasonable until you become aware that the patients most in need are often the ones least likely to improve. It also ignores the rea
72.
by derbOac 5mo ago
Very happy to see this in the news today.
73.
by derbOac 5mo ago
What's ironic to me in these discussions is how similar Ukraine was at one time to the current US administration (probably not by coincidence). Things change quickly.
74.
by derbOac 5mo ago
FWIW, I interpreted the article as saying they're not looking for vibe coding, but AI model development per se: "...In practical terms, GM is looking for people who know how to build with AI from the ground up — designing the syst
75.
by derbOac 5mo ago
https://en.wikipedia.org/wiki/Opportunity_cost
76.
by derbOac 5mo ago
Think about the anger toward Clippy. Now think about Clippy, but where feeding Clippy is a significant part of GDP, and there's a religious fervor around Clippy, especially among the older and wealthy. That's my personal impressio
77.
by derbOac 5mo ago
I used to do patient care in this area, pre GLP-1, and am extremely glad to see these medications taking off and displacing surgery. I wouldn't call the surgical interventions a scam, but their risks have routinely been understated and
78.
by derbOac 5mo ago
Thanks for that link. I've seen this photo before but never with any historical context, other than its significance as a photography milestone. That site explains the context of it as a news photo relatively well.
79.
by derbOac 5mo ago
One way to think of it — I didn't read the article in depth so this is just an example — is in terms of overall individual differences in speed and activity level. Then, you could have slower persons having increased activity relative
80.
by derbOac 5mo ago
"Quality metrics" need much more discussion and attention, in my opinion. Not a criticism of this project — it's a good idea, it just highlights the central question of "how well is this model working?" I'm not
81.
by derbOac 5mo ago
In my area it's become pretty popular, pretty broadly but especially with younger individuals.
82.
by derbOac 6mo ago
Yeah I don't think people are really fully appreciating the scope of this, because it means people would essentially have to have a passport to open a bank account. It's very dark. I tend to be libertarian about these things and f
83.
by derbOac 6mo ago
Nintendo moving from playing cards (although they still make them) to computer gaming makes much more sense to me than this.
84.
by derbOac 6mo ago
Simple but clever study. One issue is heterogeneity in effects across individuals. Even with traditional antidepressants, it's common for people to respond to one drug class but not another. So it's conceivable, just based on this
85.
by derbOac 6mo ago
FWIW, I like her videos but I usually prefer essays or blog posts in general as they're easier to scan and process at my own rate. It's not about this particular video, it's about videos in general.
86.
by derbOac 6mo ago
An "autonomous market with enough agents" is carrying a lot of weight there, like "rational actors" and "as sample size goes to infinity'.
87.
Enduring constraints on grammar revealed by Bayesian spatiophylogenetic analyses(nature.com)
2 points by derbOac 6mo ago|1 comments
88.
by derbOac 6mo ago
I wish I could read this, and I wish even more it had been open access or otherwise published in an open way — archive.today can't get behind the firewall, and I can't access it at my university (an R1 institution) either because
89.
by derbOac 6mo ago
It depends on the program, and even more so, the student and the mentor. It can also vary over time, with more direction early on in a graduate program, and less direction later. Some mentors are very directive, and basically treat students
90.
by derbOac 7mo ago
I feel like those links are more useful than the target essay. Reading through them, I wonder why CPIs aren't based on empirical correlational patterns between prices over time? Sort of like in these articles: https://iopsci

More ›