LiveThinking: efficient reasoning for real-time QA
Distill the capability, then use reinforcement learning to remove unnecessary thinking.
Y'LOG / COLLECTION
2 篇文章 · 均含中英文版本
Distill the capability, then use reinforcement learning to remove unnecessary thinking.
从 DeepSeek-R1 蒸馏到推理路径压缩,介绍 long2short DPO、质量与长度双偏好 DP-DPO 及泛化性优化实验。