Mike Knoop

speaker
927 appearances 4 recordings 1 series first heard Mar 2025 last heard 18 Nov

Mike Knoop’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months
1 · Nov OctJan 26AprJulnow

Recordings per month over the last 12 months — 1 in all, peaking in Nov 2025 with 1.

Appearances

newest first · ▶ plays the moment
Um Arc this is one of the reasons Arc V1 lasted five years was because it was it had very good design original design built into it with a public test set and a private test set.
Sure.
And that private test set really prevented folks from being able to sort of overfit on it.
Um and um so we there's actually two tracks for the Arc Prize Foundation.
We have the the contest track, this is on Kaggle, this is the like big grand prize, all the prize money is attached to this.
Uh Kaggle graciously donates a lot of compute to allow us to host there.
And on uh the grand prize is basically to get to that 85% within Kaggle's efficiency limits.
So this is about get about $50 worth of compute per submission that you send in.
Um and
And uh this is like a pretty high bar for efficiency.
And in fact, this is not an arbitrary bar.
We do think that efficiency is actually the really, really important aspect of intelligence.
Totally.
Yeah.
You know, you can brute force your way up to intelligence, but we really do want to be shooting for like human targets and efficiency for this stuff.
Yeah, but
That's the contest.
When we launched last year, uh there was a lot of demand from just the community of the world on like, hey, I okay, I get that like I can't run my big language model on this, you know, in Kaggle, but I really want to know how to frontier AI systems.
And so we launched a second track that's hosted on archprise.org where we benchmark and test all the like frontier commercial systems to showcase what are they capable of doing, basically these existence proofs that I think will eventually filter down to higher efficiency solutions and open source solutions that can be run on something like Kaggle.
Um and even there, like within this sort of same efficiency and uh you know to accuracy specs, we still haven't seen any frontier AI system beat uh
Showing 721–740 of 927 · page 37 of 47 ← Previous Next →