Ira Assent

Least but not Last: Fine-tuning Intermediate Principal Components for Better Performance-Forgetting Trade-Offs

A comprehensive analysis of the performance-forgetting trade-offs inherent in low-rank adaptation using principal components of weight matrices as initialization reveals that fine-tuning intermediate components leads to better balance and robustness to high learning rates than first (PiSSA) and last (MiLoRA) components in existing work.