行业动态2026/8/2714 次阅读A Thread-Register Decoupled GPU Execution Model for Efficient Tensor Computation来源:arxiv.org 来源: https://arxiv.org/abs/2608.19628 热度: 11 points on HackerNews来源: https://arxiv.org/abs/2608.19628热度: 11 points on HackerNews