Hardware-Aware Features for CUTLASS Kernel Selection

cs.LG, cs.PF

Submitted: 2026-09-28

Updated: 2026-09-28

Comments: 20 pages, 19 figures

Code: https://github.com/spcl/cutlassselector

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Learning from distinctive candidates to optimize reduced-precision convolution program on tensor cores
• Microbenchmark-Driven Analytical Performance Modeling Across Modern GPU Architectures
• Understanding GEMM Performance and Energy on NVIDIA Ada Lovelace: A Machine Learning-Based Analytical Approach
• KernelBand: Steering LLM-based Kernel Optimization via Hardware-Aware Multi-Armed Bandits
• MetaTune: Meta-Learning Based Cost Model for Fast and Efficient Auto-tuning Frameworks
• Optimal Software Pipelining and Warp Specialization for Tensor Core GPUs
• tritonBLAS: Triton-based Analytical Approach for GEMM Kernel Parameter Selection
• WaveTune: Wave-aware Bilinear Modeling for Efficient GPU Kernel Auto-tuning
• Hexcute: A Compiler Framework for Automating Layout Synthesis in GPU Programs

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks