Shears: Unstructured Sparsity with Neural Low-rank Adapter Search
This work addresses parameter-efficient fine-tuning for large language models, offering an incremental improvement in efficiency and compression.
The paper tackles the problem of efficiently fine-tuning and compressing large language models by introducing Shears, which combines sparsity with a neural low-rank adapter search algorithm, achieving high sparsity levels with minimal accuracy drop using a single GPU in a few hours.
Recently, several approaches successfully demonstrated that weight-sharing Neural Architecture Search (NAS) can effectively explore a search space of elastic low-rank adapters (LoRA), allowing the parameter-efficient fine-tuning (PEFT) and compression of large language models. In this paper, we introduce a novel approach called Shears, demonstrating how the integration of cost-effective sparsity and a proposed Neural Low-rank adapter Search (NLS) algorithm can further improve the efficiency of PEFT approaches. Results demonstrate the benefits of Shears compared to other methods, reaching high sparsity levels while improving or with little drop in accuracy, utilizing a single GPU for a pair of hours.