Enabling Lightweight Fine-tuning for Pre-trained Language Model Compression based on Matrix Product Operators

Open in new window