作者位次 #3/11ACL 2026Findings· Findings of ACL
C o M o L : Efficient Mixture of L o RA Experts via Dynamic Core Space Merging
Jie Cao · Zhenxuan Fan · Zhuonan Wang · Tianwei Lin · Ziyuan ZHAO · Rolan Yan · Wenqiao Zhang · Feifei Shao · Hongwei Wang · Jun Xiao · Siliang Tang
摘要
Large language models (LLMs) achieve remarkable performance on diverse downstream and domain-specific tasks via parameter-efficient fine-tuning (PEFT). However, existing PEFT methods, particularly MoE-LoRA architectures, suffer from limited parameter efficiency and coarse-grained adaptation due t...