Beyond Domain-Routed Supervision: Cross-Teacher Guarding for Multi-Teacher On-Policy Distillation
🔗 https://huggingface.co/papers/2610.13521 ↗Road MOPD reduces gradient conflicts in multi-teacher distillation by using cross-teacher agreement for token-level credit assignment and auxiliary gradients to constrain conflicting updates.
內容為收集當日抓取的摘要原文。