improve AI discourse about 5% just by renaming evals accurately Humanity's Last Exam: PubQuizFromHell MATH: RemedialMath FrontierMath: QuarterFrontierMath SWE-Bench: DjangoBench MMLU Virology:…
Posters: gleech. gleech
3 likes recorded by the source corpus.
Topics: AI & LLMs, Epistemics