AI Hub
All papers
Training

Textbooks Are All You Need

Microsoft·June 20, 2023

Suriya Gunasekar, Yi Zhang, Jyoti Aneja

View on arXiv

TL;DR

Shows that small models trained on curated, textbook-quality data can rival much larger ones (the Phi line).

Why it matters

Made the case that data quality can substitute for raw scale, shaping the small-model wave.

Related terms