BROWSE
Detail
Conference Paper
PC-LoRA: Low-Rank Adaptation for Progressive Model Compression with Knowledge Distillation
Cited - time in 
Share 
Share
- Authors
- Injoon Hwang, HaeWon Park, Youngwan Lee, Jooyoung Yang, SunJae Maeng
- Issue Date
- 2024-06
- Citation
- Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2024, pp.1-8
- Language
- English
- Type
- Conference Paper
- 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.
- KSP Keywords
- Fine-tuning, Knowledge Distillation, Language Model, Model compression, Progressive compression, Rank adaptation, empirical analysis, low rank, training process, various models