Repetition over Diversity: High-Signal Data Filtering for Sample-Efficient German Language Modeling Paper • 2604.28075 • Published 7 days ago • 18
Pre-Training Curriculum for Multi-Token Prediction in Language Models Paper • 2505.22757 • Published May 28, 2025
Repetition over Diversity: High-Signal Data Filtering for Sample-Efficient German Language Modeling Paper • 2604.28075 • Published 7 days ago • 18