---
id: 20260930-T0-06
title: "OMP-MoE：用正交匹配追踪高效剪枝MoE大模型专家"
title_en: "OMP-MoE: Orthogonal Matching Pursuit for Efficient Expert Pruning in MoE LLMs"
url: https://ai.daily.yangsir.net/daily/20260930-T0-06
issue_date: 2026-09-30
publish_date: 2026-09-29T04:00:00.000Z
category: research
source_name: "arXiv cs.LG (ML)"
source_url: https://arxiv.org/abs/2609.31631
---

# OMP-MoE：用正交匹配追踪高效剪枝MoE大模型专家

研究者提出OMP-MoE方法，利用正交匹配追踪（Orthogonal Matching Pursuit）对混合专家（MoE）大语言模型进行专家剪枝。MoE架构虽能高效扩展模型规模，但巨大的内存需求给部署带来挑战。现有剪枝方法要么搜索成本过高，要么忽视专家间的动态交互关系。OMP-MoE旨在以更低计算开销完成专家筛选，降低MoE模型的部署内存门槛，使大规模MoE模型在资源受限环境中更易落地。

## English Version

**OMP-MoE: Orthogonal Matching Pursuit for Efficient Expert Pruning in MoE LLMs**

Researchers propose OMP-MoE, a method that applies Orthogonal Matching Pursuit to prune experts in Mixture-of-Experts (MoE) large language models. While MoE architectures enable efficient scaling, their massive memory requirements pose deployment challenges. Existing pruning methods either incur prohibitive search costs or neglect dynamic inter-expert interactions. OMP-MoE aims to reduce computational overhead during expert selection, lowering memory barriers for deploying large-scale MoE models in resource-constrained environments.

---

**来源**：[arXiv cs.LG (ML)](https://arxiv.org/abs/2609.31631)

**详情页**：https://ai.daily.yangsir.net/daily/20260930-T0-06

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*