Skill-Conditioned Gated Self-Distillation for LLM Reasoning 文章

ArXiv CS.CL2026-05-28NEWSen作者: Jiazhen Huang, Xiao Chen, Xiao Luo, Yong Dai, Senkang Hu, Yuzhi Zhao
查看原文 →
关系图谱

Skill-Conditioned Gated Self-Distillation for LLM Reasoning · 相关技术

相关技术

LLMteacher-side privileged informationSkill-Conditioned Gated Self-DistillationOn-policy self-distillation