Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs Paper • 2403.20041 • Published Mar 29, 2024 • 34