American Mathematics Competitions, every mention
4 scenes · ← back to American Mathematics Competitions
tap a year for its mentions
every year anyone Stanislas Polu 2Nathan Lambert 1Michelle Pokrass 1Charles Packer 1
Verbatim, from the transcripts: the passages where American Mathematics Competitions comes up
Sleep-Time Compute — Letta AI (Charles Packer, Charlie Snell, Kevin Lin)
- ▶ 14:57 Charles Packer I think in this case, you know, with these like GSM, AK style questions or like Amy style, there's going to be a limit.
GPT 4.1: The New OpenAI Workhorse
- ▶ 27:45 Michelle Pokrass So Amy, GPQA, stuff like that, you'll see the reasoning models do much better.
Agents @ Work: Dust.tt — with Stanislas Polu
- ▶ 7:54 Stanislas Polu And I think it was the low end part of the mass benchmark at the time, because that mass benchmark includes AMC problems, AMC eight times 10, 12. 2 times in the scene
The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
- ▶ 1:00:06 Nathan Lambert They essentially listed their kind of bogus evaluations, because it's a hilarious table, because it's like LSAT AP exams, and then like AMC-X and AMC-X are like,