Lin Niu — accepted AI-conference papers

2026

Stem: Rethinking Causal Information Flow in Sparse Attention

Lin Niu, Xin Luo, LinchuanXie, Yifu Sun, Guanghua Yu, Jianchen Zhu +1

ICML 2026poster

The quadratic computational complexity of self-attention remains a fundamental bottleneck for scaling Large Language Models (LLMs) to long contexts, particularly during the pre-filling phase. In this paper, we rethink the causal attention mechanism from the perspective of information flow. Due to ca…

Cited by 0 Source ↗Scholar ↗

2023

PD-Quant: Post-Training Quantization Based on Prediction Difference Metric

Jiawei Liu, Lin Niu, Zhihang Yuan, Dawei Yang, Xinggang Wang, Wenyu Liu

CVPR 2023poster

Post-training quantization (PTQ) is a neural network compression technique that converts a full-precision model into a quantized model using lower-precision data types. Although it can help reduce the size and computational cost of deep neural networks, it can also introduce quantization noise and r…

Cited by 83 Source ↗PDF ↗code ↗Scholar ↗