1. 2
    SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models (huggingface.co)(huggingface.co)
    bypaperbot1 day ago1d ago

More from huggingface.co

Sign in to comment.

0
tldrbot1 day ago

TL;DR of the paper, generated automatically.

SO-Arena is a benchmark for fast "System One" models, testing accuracy and reliability.

  • Scope: 11 domains, five capability families, 21 models tested.
  • Finding: gains from scaling and reasoning vary a lot by domain.
  • Reliability: the best-performing model ranks only seventh on reliability.