Retargetable Compilation for Tensor Kernels via a Unified Layout Abstraction
cs.PL
Submitted: 2026-11-10
Updated: 2026-11-10
Code: https://github.com/NVIDIA/cutlass
Terminology
Sources
- Tiramisu: A Polyhedral Compiler for Expressing Fast and Portable Code
- FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning
- Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs
- TensorIR: An Abstraction for Automatic Tensorized Program Optimization
- nncase: An End-to-End Compiler for Efficient LLM Deployment on Heterogeneous Storage Architectures
- Relay: A High-Level Compiler for Deep Learning
- Glow: Graph Lowering Compiler Techniques for Neural Networks
- Tensor Program Optimization with Probabilistic Programs
- Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
- TileLang: A Composable Tiled Programming Model for AI Systems
- SparseTIR: Composable Abstractions for Sparse Compilation in Deep Learning
Related papers
- The Best of Times, the Worst of Times: Moment-Based Analysis of Probabilistic Cost Structures
- Formal Verification of an Out-of-Order Multiprocessor against an In-Order Weak-Memory ISA
- Fuss-free cumulative universes: theory and practice
- High-Level Synthesis of Efficient Pipelines with Visibility Control
- Model Checking Probabilistic Multiparty Session Types
- A Type System for Communication-Safe Hardware Design