1 paper
Yong Liu, Di Fu, Yang Luo +4
We introduce Post-Optimization Model Edit (POME), a new algorithm that enhances the performance of fine-tuned large language models using only their pretrained and fine-tuned check…