TY  - RPRT
TI  - Fast and Accurate: Improving Personalized LLM Performance by Semantic Concentration Inspired Workload Split between RAG and Fine-Tuning
AU  - Xingke Yang
AU  - Sicong Li
AU  - Lingfeng Yao
AU  - Rui Chen
AU  - Changqing Luo
AU  - Xin Fu
AU  - Miao Pan
PY  - 2026
UR  - https://arxiv.org/abs/2610.15921
ID  - 2610.15921
ER  -