Commenter pushes back on perplexity claim: math gains likely came from RL, not pretraining

stochasticchasm · x · 2026-11-23

stochasticchasm challenges the framing that upgrading to a model with log-perplexity 0.76 produced groundbreaking math results: such numbers are hard to interpret without knowing the dataset, context length, and other details. He argues the explosive math capability gains were primarily driven by RL rather than pretraining.

Related event: OpenAI's Perplexity Narrative Questioned: RL, Not Pretraining, Drove Math Breakthrough(2 posts)→

Original post →

More from Models

Models channel →