MLIR: A Compiler Infrastructure for the End of Moores Law, Chris Lattner, Mehdi Amini, River Riddle, Albert Cohen, Alan Mycroft, Chris Leary, John S. Oliver, Andrew Davis, Edward R. Lafferty, Vinod Grover, John Kew, Jason Furman, Stephan Herhut, Paulius Stankaitis, Stephen Neuendorffer, Andy Lin, Matthew Allman, James Manson, Andy Keep, N. S. Jayasena, 2021Proceedings of the ACM on Programming Languages (PACMPL), Vol. 5 (ACM)DOI: 10.1145/3434384 - 介绍了MLIR,一个用于构建现代ML编译器,具备多级中间表示的框架,对理解渐进式降低过程至关重要。
TVM: An Automated End-to-End Optimizing Compiler for Deep Learning, Tianqi Chen, Thierry Moreau, Ziheng Jiang, Lianmin Zheng, Eddie Yan, Haichen Shen, Meghan Cowan, Leyuan Wang, Yuwei Hu, Luis Ceze, Carlos Guestrin, Arvind Krishnamurthy, 201813th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18) (USENIX Association) - Apache TVM的奠基性论文,该文介绍了节中讨论的开源ML编译器和运行时栈,详细阐述了其架构、Relay IR和自动调优功能。