PC-LoRA: Low-Rank Adaptation for Progressive Model Compression with Knowledge Distillation

TL;DR

PC-LoRA introduces a progressive method that combines model compression and fine-tuning by gradually replacing pre-trained weights with low-rank adapters, achieving high compression rates in vision and language models.

Contribution

It presents a novel progressive approach that removes pre-trained weights during training, enabling simultaneous model compression and fine-tuning with low-rank adapters.

Findings

Achieves over 94% parameter compression in vision models.

Attains over 93% parameter compression in language models.

Reduces FLOPs significantly while maintaining performance.

Abstract

Low-rank adaption (LoRA) is a prominent method that adds a small number of learnable parameters to the frozen pre-trained weights for parameter-efficient fine-tuning. Prompted by the question, ``Can we make its representation enough with LoRA weights solely at the final phase of finetuning without the pre-trained weights?'' In this work, we introduce Progressive Compression LoRA~(PC-LoRA), which utilizes low-rank adaptation (LoRA) to simultaneously perform model compression and fine-tuning. The PC-LoRA method gradually removes the pre-trained weights during the training process, eventually leaving only the low-rank adapters in the end. Thus, these low-rank adapters replace the whole pre-trained weights, achieving the goals of compression and fine-tuning at the same time. Empirical analysis across various models demonstrates that PC-LoRA achieves parameter and FLOPs compression rates of 94.36%/89.1% for vision models, e.g., ViT-B, and 93.42%/84.2% parameters and FLOPs compressions for language models, e.g., BERT.