Home » Authors » Tianqi Chen, Xunchao Li, Linchuan Xie, Guanghua Yu, Jianchen Zhu, Peisong Wang, Jian Cheng

Tianqi Chen, Xunchao Li, Linchuan Xie, Guanghua Yu, Jianchen Zhu, Peisong Wang, Jian Cheng

Public reference bibliography and cataloged PDF materials authored or published by Tianqi Chen, Xunchao Li, Linchuan Xie, Guanghua Yu, Jianchen Zhu, Peisong Wang, Jian Cheng.

Found 1 document(s) Page 1 of 1

RAQ-MLA: Role-Aware Quantization for Multi-Head Latent Attention Caches

Year: 2026 • Source: arxiv.org • Listed PDF

Multi-Head Latent Attention (MLA) has been adopted by recent frontier large language models. Instead of per-head keys and values, MLA caches a compressed KV latent that is up-projected into keys and values at decode time...