Distillation of Large Language Models via Concrete Score Matching 文章

ArXiv CS.AI2026-06-02NEWSen作者: Yeongmin Kim, Donghyeok Shin, Mina Kang, Byeonghu Na, Il-Chul Moon
查看原文 →
关系图谱

Distillation of Large Language Models via Concrete Score Matching · 相关技术

相关技术

logit shift invarianceautoregressivediscrete score-matchingsoftmaxConcrete Score Distillationlarge language modelsDirect Logit DistillationKnowledge Distillation