← Search

Su Zhao

2 accepted papers

2025

Compress Large Language Models via Collaboration Between Learning and Matrix Approximation

NeurIPS 2025poster

Sparse and low-rank matrix composite approximation has emerged as a promising paradigm for compressing large language models (LLMs), offering a more flexible pruning structure than conventional methods based solely on sparse matrices. The significant variation in weight redundancy across layers, alo…

Cited by 0SourceScholar
2025

Efficient Representativeness-Aware Coreset Selection

NeurIPS 2025poster

Dynamic coreset selection is a promising approach for improving the training efficiency of deep neural networks by periodically selecting a small subset of the most representative or informative samples, thereby avoiding the need to train on the entire dataset. However, it remains inherently challen…

Cited by 0SourceScholar