Jinglin Liu

Papers on Lattice

Total citations

Topics

h-index

Publication activitypapers/week, last 8 weeks

Research focus

Architecture Design (Transformers, SSMs, MoE) (1)Distributed Systems & Hardware (1)Inference & Quantization (1)

Frequent co-authors

Chang Han (1)Yijie Hu (1)Jingling Liu (1)

Papers (1)

Mar 9, 2026

2d ago

EAGLE-Pangu: Accelerator-Safe Tree Speculative Decoding on Ascend NPUs

Tree speculative decoding can achieve up to 2.46x speedup on Ascend NPUs, but only if you carefully manage the branch/commit cache and eliminate undefined negative indices.

Chang Han, Yijie Hu, Jinglin Liu +1

Architecture Design (Transformers, SSMs, MoE)Distributed Systems & Hardware Inference & Quantization

Search

Jinglin Liu

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)