PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zou, Lancheng, Yin, Shuo, Pei, Zehua, Ho, Tsung-Yi, Farnia, Farzan, Yu, Bei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908587630002176
author Zou, Lancheng
Yin, Shuo
Pei, Zehua
Ho, Tsung-Yi
Farnia, Farzan
Yu, Bei
author_facet Zou, Lancheng
Yin, Shuo
Pei, Zehua
Ho, Tsung-Yi
Farnia, Farzan
Yu, Bei
contents Channel permutation is a powerful technique for enhancing the accuracy of N:M sparse models by reordering the channels of weight matrices to prioritize the retention of important weights. However, traditional channel permutation methods rely on handcrafted quality metrics, which often fail to accurately capture the true impact of pruning on model performance. To address this limitation, we propose PermLLM, a novel post-training pruning framework that introduces learnable channel permutation (LCP) for N:M sparsity. LCP leverages Sinkhorn normalization to transform discrete permutation matrices into differentiable soft permutation matrices, enabling end-to-end optimization. Additionally, PermLLM incorporates an efficient block-wise channel permutation strategy, which significantly reduces the number of learnable parameters and computational complexity. PermLLM seamlessly integrates with existing one-shot pruning methods to adaptively optimize channel permutations, effectively mitigating pruning-induced errors. Extensive experiments on the LLaMA series, Qwen, and OPT models demonstrate that PermLLM achieves superior performance in optimizing N:M sparse models. The code is available at https://github.com/lanchengzou/PermLLM.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10136
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models
Zou, Lancheng
Yin, Shuo
Pei, Zehua
Ho, Tsung-Yi
Farnia, Farzan
Yu, Bei
Machine Learning
Artificial Intelligence
Channel permutation is a powerful technique for enhancing the accuracy of N:M sparse models by reordering the channels of weight matrices to prioritize the retention of important weights. However, traditional channel permutation methods rely on handcrafted quality metrics, which often fail to accurately capture the true impact of pruning on model performance. To address this limitation, we propose PermLLM, a novel post-training pruning framework that introduces learnable channel permutation (LCP) for N:M sparsity. LCP leverages Sinkhorn normalization to transform discrete permutation matrices into differentiable soft permutation matrices, enabling end-to-end optimization. Additionally, PermLLM incorporates an efficient block-wise channel permutation strategy, which significantly reduces the number of learnable parameters and computational complexity. PermLLM seamlessly integrates with existing one-shot pruning methods to adaptively optimize channel permutations, effectively mitigating pruning-induced errors. Extensive experiments on the LLaMA series, Qwen, and OPT models demonstrate that PermLLM achieves superior performance in optimizing N:M sparse models. The code is available at https://github.com/lanchengzou/PermLLM.
title PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.10136