arXiv Machine Learning
AttnFuse: A Composable DSL for Compiling Attentions to Fused GPU Kernels
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Sep 16
Ave: Guiding Agentic GPU Optimization Using Data-Flow Invariants
arXiv:2604.18616v2 Announce Type: replace-cross Abstract: LLM coding agents can generate correct GPU kernels, but their performance still trails expert libraries. Reaching peak throughput requires co...
arXiv Machine Learning
Oct 3
Fast Polynomial Transcendentals for LLMs
The paper investigates using short polynomial approximations to accelerate special‑function operations in large language models on NVIDIA Blackwell GPUs. By replacing native sigmoid, tanh, and SiLU with degree‑3 or degree‑4 bfloat16 programs, the authors achieve up to 2.19× speed‑ups in isolated FP16 benchmarks and modest training‑step throughput gains (2.7–8.0%) across four integration tasks. The study also evaluates model behavior, finding negligible training‑loss differences within 100 billion tokens.
arXiv AI
Jun 9
Attention at the Theoretical Minimum: A Mathematics of Arrays Framework for Memory-Optimal Transformer Kernels
arXiv:2606. 07713v1 Announce Type: cross Abstract: The attention mechanism is the dominant computational bottleneck in modern transformer-based AI.
arXiv Machine Learning
Aug 4
Nova: An End-to-End MLIR Compiler for Deep Learning
arXiv:2608. 00029v1 Announce Type: cross Abstract: The performance of deep learning models at scale relies heavily on how effectively high-level mathematical operations are mapped to underlying physical hardware.
arXiv AI
Sep 25
KernelOPT: Dispatch-Aware Agentic Search for GPU Kernel Optimization
KernelOPT is a multi‑agent system that optimizes GPU kernels generated by compilers like PyTorch Inductor by treating compiled models as structured artifacts. It preserves vendor library calls and focuses on Triton sub‑kernels, using five profiling‑guided LLM agents and a four‑gate verification cascade to ensure correctness and performance before re‑stitching the model. On 250 KernelBench problems, KernelOPT achieves geometric mean speedups of 1.40×, 1.15×, and 1.07× over torch.compile at three optimization levels.
arXiv AI
Aug 24
HIERA: Workload-Aware Planning Across Implementation Spaces for GPU Kernel Optimization
arXiv:2608.21157v1 Announce Type: cross Abstract: High-performance GPU kernels underpin modern deep learning and scientific computing. As workloads become increasingly diverse and GPU hardware evolve...