szmerdi

4 ms·

1.
by szmerdi 3y ago
I like the initiative and the large scale collaboration but in terms of quality, my observation on one of the newly added languages (an underrepresented one) is it outputs dramatically grammatically gibberish! As though it hasn't been
2.
by szmerdi 3y ago
Each year, national math olympiad committees submit their toughest problems to the IMO. A select group makes the "shortlist," a pool of ~32 questions (8 per geometry, algebra, number theory, combinatorics). From this pool, the fin
3.
by szmerdi 3y ago
As a former problem designer for IMO and similar contests, I deeply enjoyed reading this paper. At the same time, I'd like to point out that it was clear Geometry had to be the first topic to give up against AI (i.e., smart knowledge a
4.
by szmerdi 3y ago
Similar project with enough technical details: https://asmirnov.xyz/doppelganger
5.
by szmerdi 3y ago
You're right - we don't know yet but, as mentioned in other replies, 20B is a reasonable guestimate that would make the point about unfair comparison.
6.
by szmerdi 3y ago
Good initiative but it's not a fair comparison IMO. The Gemini Ultra is the one that should've been compared to GPT-4. And, yes, I know it's not available behind an API yet. But does it justify comparing a < 20 billion par
7.
by szmerdi 3y ago
I enjoy playing around with https://ideogram.ai much more. Correct spelling was always there and you can mix and match with others' prompts: the image generation experience is a collective creative activity. Bing is despera