Niklas Muennighoff

This research focuses on optimizing language model training and predicting their real-world performance.

This article discusses issues and best practices for evaluating language models.

SEACrowd aims to improve AI representation for Southeast Asian languages and cultures.

A new method improves the selection of data mixtures for language model training.

Discover how vocabulary size influences the performance of large language models.

Precision impacts the effectiveness and cost of language model training.