Fetching the paper…

SPT: Fine-Tuning Transformer-based Language Models Efficiently with Sparsification · Around