Xe-Forge: Multi-Stage LLM-Powered Kernel Optimization for Intel GPU 文章

ArXiv CS.AI2026-05-27NEWSen作者: Marcin Spoczynski, Daniel Fleischer, Moshe Berchansky, Gabriela Ben-Melech Stan, Shira Guskin, Weilin Xu, Adam Siemieniuk, Alexander Heinecke
人工智能
查看原文 →
关系图谱

Xe-Forge: Multi-Stage LLM-Powered Kernel Optimization for Intel GPU · 相关技术

相关技术

TTEODE通用表达式语言UCTStraight-Through EstimatorReferring expression comprehension (REC)PETLMMKernelHISForEffort Metric AttentionEFICamouflaged object detectionARGANNAgentICEGPU