GQLA: Group-Query Latent Attention for Hardware-Adaptive Large Language Model Decoding 文章

ArXiv CS.AI2026-05-28NEWSen作者: Fanxu Meng
查看原文 →
关系图谱

GQLA: Group-Query Latent Attention for Hardware-Adaptive Large Language Model Decoding · 相关技术

相关技术

Grouped-Query AttentionMulti-Query Attentiontensor parallelismGQAMQAGroup-Query Latent AttentionMulti-Token PredictionMulti-head Latent Attention