I wonder how well humans would do in this chart.
21,220 karma · joined July 8, 2021
For every problem you can't solve, there's a simpler problem that you also can't solve.
Nintendo's victory means European file-sharing companies must now remove illegal copies of games when asked to do so
Suppose I'm interested in representing a Group as matrices over the complex numbers. There are usually many ways of doing this. Each one of them has a so-called character, which is like fingerprint of such a representation.
Along another line, it has been known that all groups contain large subgroup having an order which is a power of a prime--call it P. This group in turn has a normalizer in which P is normal--call it N(P).
The surprising thing is that the number of characters of G and of N(P)--which is is only a small part of G--is equal.
*technical note in both cases we exclude representation the degree of which is a multiple of p.
It would be pity to loose access to o3 though. (checked this morning, and I still had access)
Curiously, Andrew Krapivin, the genious undergrad in the article, is not one of the authors.
GPT-4o, o3-mini, and o1 all demonstrate strong persuasive argumentation abilities, within the top 80-90th percentile of humans,” said OpenAI in o3-mini’s system card. “Currently, we do not witness models performing far better than humans, or clear superhuman performance.
The outlet’s sources said Microsoft security researchers detected that large amounts of data were being exfiltrated through OpenAI developer accounts in late 2024, which the company believes are affiliated with DeepSeek. OpenAI told the Financial Times that it found evidence linking DeepSeek to the use of distillation (...)
In fact, it is exactly the point of the article that more access was used than is provided in the API. Distillation is not possible with API access.
"Induced demand": The highway is expanded such that it only takes me 15 minutes to get downtown instead of 45, so I go downtown more.
Jevson paradox: The highway is expanded such that it only takes 15 minutes to get from the suburbs to my office instead of 45, so I move to the suburbs.
So one is movement along the demand curve, while the other is a movement of the demand curve.
https://www.reddit.com/r/badeconomics/comments/oo2s48/that_o...
Dutch anti-piracy group BREIN announced that one of the largest Dutch LLMs ‘GEITje-7B‘ was taken offline as a result of their efforts.
This LLM was trained on ‘Gigacorpus’ a dataset of books and texts previously targeted by BREIN, including a vast collection of Dutch texts and books, some of which contained copyrighted material sourced from the shadow library LibGen.
As far as I know, this is the first time LLMs are taken down based on their training on copyrighted materials.
That's nonsense. I'm sure there thinking of RSA, but that needs secret prime numbers. So easy-to-remember is pretty much the opposite of one want. Also they are way to big. 2048 bit RSA needs two 300 digit prime numbers.
All this additional regulation, doesn't change a wit about the models and their impact, but does keep open source, start ups, academia out of the game.
- You can hardly use the writing in an essay, as it is replete with listicles. You'd have to completely rewrite the stuff anyway. Moreover, he know he'll be tested on this stuff, and yeah that does seem to have an impact.
- Even using ChatGpt for explaining stuff is hardly used. For sure, I use it much much more often than he does. Also here it seems that it is hardly worth it, as the ChatGpt answer will differ too much from what is expected on the test. He much prefers to study with last years' tests--and the accepted answers-- than to use ChatGpt.
https://pubmed.ncbi.nlm.nih.gov/39550055/
What I find cool, is that the study involved an open source data set with brain imaging data of a 1000 individuals (the Human Connectome Project Young Adult dataset). Anyone could have downloaded the data set, done the analysis, and scored a publication in Nueroimage. (assuming, you are smart, and have the drive of course..)