Episode Details
Back to Episodes
s1: simple test time scaling
Published 1 year, 5 months ago
Description
- Test-time scaling improves language model performance using extra compute
- A dataset of 1,000 questions was curated for validation
- Budget forcing controls compute by managing the model's reasoning process
- The model outperformed o1-preview by up to 27% on math questions
- The model and data are open-source for public access