AgentKernelArena: Generalization-Aware Benchmarking of GPU Kernel Optimization Agents

arXiv:2605.16819v1 Announce Type: cross Abstract: GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing high-performance kernels still requires substantial low-level expertise. Recent AI coding agents can iteratively read code, invoke compilers and profilers, and refine implementations, yet existing kernel benchmarks evaluate single LLM calls rather than full agent workflows, and none include both kernel-to-kernel optimization and unseen-configuration