Researcher, ByteDance
2 papers at NeurIPS 2025
We proposed a new data selection method for pretraining multilingual Large Language Models