Study: Larger models tolerate more data repetition during pretraining

StanfordAILab · x · 2026-08-22

StanfordAILab retweeted research on data repetition in pretraining. The study finds that at fixed tokens/parameters, larger models can tolerate more data repetition. Larger models and shorter LR decay also help. This suggests repeating high-quality data might be viable when unique data is scarce.

Original post →

More from Research

Research channel →