PrunePath: Towards Highly Structured Sparse Language Models 事件

PrunePath: Towards Highly Structured Sparse Language Models arXiv:2605.28283v1 Announce Type: new Abstract: Feed-forward networks (FFNs) dominate the parameter count and computation of modern language models, yet existing pruning methods often struggle to convert sparsity into hardware-friendly inference efficiency gains. We introduce \textbf{PrunePath}, a budget-adaptive structured sparsification framework for FFN layers. Built on MoEfication, PrunePath replaces independent expert-wise thresho