4 papers
Yggdrasil: Bridging Dynamic Speculation and Static Runtime for Latency-Optimal Tree-Based LLM Decoding
Yue Guan, Changming Yu, Shihan Fang +8
Speculative decoding improves LLM inference by generating and verifying multiple tokens in parallel, but existing systems suffer from suboptimal performance due to a mismatch betwe…
VQ-LLM: High-performance Code Generation for Vector Quantization Augmented LLM Inference
Zihan Liu, Xinhao Luo, Junxian Guo +11
In this work, we design and implement VQ-LLM, an efficient fused Vector Quantization (VQ) kernel generation framework. We first introduce a software abstraction called codebook cac…
An Efficient Private GPT Never Autoregressively Decodes
Zhengyi Li, Yue Guan, Kang Yang +5
The wide deployment of the generative pre-trained transformer (GPT) has raised privacy concerns for both clients and servers. While cryptographic primitives can be employed for sec…
M-ANT: Efficient Low-bit Group Quantization for LLMs via Mathematically Adaptive Numerical Type
Weiming Hu, Haoyan Zhang, Cong Guo +7
Large language models (LLMs) are one of the most important killer computer applications. The recent algorithmic advancement proposes a fine-grained group-wise quantization for LLMs…