Fast Autoscheduling for Sparse ML Frameworks

Bobby Yan; Alexander J Root; Trevor Gale; David Broman; Fredrik Kjolstad

doi:10.1109/CGO68049.2026.11394842

CGO · February 2026

Fast Autoscheduling for Sparse ML Frameworks

Bobby Yan Alexander J Root Trevor Gale David Broman Fredrik Kjolstad

IEEE/ACM International Symposium on Code Generation and Optimization

Read PDF DOI

Topics Scheduling & autotuning Sparse tensor algebra

Related software

Scorch: An Optimized Sparse PyTorch A library that adds sparse tensors and JIT-compiled kernels to PyTorch with a compatible API, automating loop ordering, tiling, and format inference for sparse machine-learning workloads.

Cite this paper

Citation

Bobby Yan, Alexander J Root, Trevor Gale, David Broman, and Fredrik Kjolstad. Fast Autoscheduling for Sparse ML Frameworks. CGO, 2026.

BibTeX

@inproceedings{yan2026fast,
  title     = {Fast Autoscheduling for Sparse ML Frameworks},
  author    = {Bobby Yan and Alexander J Root and Trevor Gale and David Broman and Fredrik Kjolstad},
  booktitle = {IEEE/ACM International Symposium on Code Generation and Optimization},
  year      = {2026},
  month     = {feb},
  doi       = {10.1109/CGO68049.2026.11394842},
  url       = {https://compilers.stanford.edu/publications/scorch.pdf}
}

All publications