错配·破环(TIDE)
TIDE 揭穿 on-policy distillation 的低 KL 假象:用两路 mismatch 校正,让强错配下的推理从重复循环回到更短、更有答案的轨道。
0:00 / 4:21

TIDE 揭穿 on-policy distillation 的低 KL 假象:用两路 mismatch 校正,让强错配下的推理从重复循环回到更短、更有答案的轨道。
arxiv.org
arxiv.org
This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.