原生稀疏注意力内核迎来高效替代方案，提速明显且内存友好，值得关注。
arXiv:2508.18224v3 Announce Type: replace-cross Abstract: Recent advances in sparse attention mechanisms have demonstrated strong potential for reduci…
多阶段LLM流水线自动优化Intel GPU内核，释放硬件潜力，开辟AI驱动编译器新方向
arXiv:2605.26118v1 Announce Type: cross Abstract: Porting deep learning algorithms to new hardware accelerators requires developers to repeatedly appl…
自动发现并优化GPU内核，实现34.93倍平均加速，MLSys 2026竞赛开源项目。
Article URL: https://github.com/Dogacel/auto-gpu-kernel Comments URL: https://news.ycombinator.com/item?id=48274965 Points: 1 # Comments: 0
首个泛化感知基准评估GPU内核优化代理，推动自动调优技术落地。
arXiv:2605.16819v1 Announce Type: cross Abstract: GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing hi…