Meta-rater: A Multi-dimensional Data Selection Method for Pre-training Language Models

Documenti analoghi

• Topic Over Source: The Key to Effective Data Mixing for Language Models Pre-training
  di: Peng, Jiahui, et al.
  Pubblicazione: (2025)
• Efficient Pretraining Data Selection for Language Models via Multi-Actor Collaboration
  di: Bai, Tianyi, et al.
  Pubblicazione: (2024)
• Harnessing Diversity for Important Data Selection in Pretraining Large Language Models
  di: Zhang, Chi, et al.
  Pubblicazione: (2024)
• VADE: Variance-Aware Dynamic Sampling via Online Sample-Level Difficulty Estimation for Multimodal RL
  di: Hu, Zengjie, et al.
  Pubblicazione: (2025)
• Multi-Step Visual Reasoning with Visual Tokens Scaling and Verification
  di: Bai, Tianyi, et al.
  Pubblicazione: (2025)