Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe Paper โข 2604.13016 โข Published 9 days ago โข 85
UltraData Collection Ultra Scale, Ultra Quality, Ultra Coverage โข 10 items โข Updated 5 days ago โข 81
InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation Paper โข 2509.24663 โข Published Sep 29, 2025 โข 16