Building Multilingual Bridges: Data Mixing as the Pillar of Generalization for In-Language Reasoning
Researchers train Tiny Aya L2-Thinker, a 3.35B model achieving over 93% in-language reasoning across 60 languages via multilingual data mixing.
1 stories, newest first · one card per story
Researchers train Tiny Aya L2-Thinker, a 3.35B model achieving over 93% in-language reasoning across 60 languages via multilingual data mixing.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views