3 papers
cs.LG2025
Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References
Hongzheng Chen, Bin Fan, Alexander Collins +8
Modern GPUs feature specialized hardware units that enable high-performance, asynchronous dataflow execution. However, the conventional SIMT programming model is fundamentally misa…
cs.PL2025
Modeling Layout Abstractions Using Integer Set Relations
Somashekaracharya G Bhaskaracharya, Aravind Acharya, Bastian Hagedorn +1
Modern deep learning compilers rely on layout abstractions to manage the complex mapping between logical tensor structures and physical memory arrangements. CuTe layouts and Triton…
cs.PL2024
Pattern Matching in AI Compilers and its Formalization (Extended Version)
Joseph W. Cutler, Alex Collins, Bin Fan +2
PyPM is a Python-based domain specific language (DSL) for building rewrite-based optimization passes on machine learning computation graphs. Users define individual optimizations b…