Thanos通过动态分块剪枝，高效压缩大模型且保持精度。
Thanos: A Block-wise Pruning Algorithm for Efficient Large Language Model Compression
- 采用自适应掩码的分块剪枝，支持灵活稀疏模式
- 在结构化与非结构化剪枝上均达当前最佳效果
- 适合资源受限场景下的大模型部署
本文提出Thanos，一种新型权重剪枝算法，通过移除冗余权重降低大语言模型（LLMs）的内存占用并提升计算效率，同时保持模型精度。Thanos引入基于块的剪枝策略与自适应掩码，可动态调整权重重要性，支持灵活的稀疏模式（如$n:m$稀疏）和利于硬件加速的结构化格式。实验表明，Thanos在结构化剪枝中表现领先，且在非结构化剪枝中优于现有方法。该方法为资源受限环境中的大模型部署提供了高效且可适配的压缩方案。
原文摘要 · Abstract (English)
This paper presents Thanos, a novel weight-pruning algorithm designed to reduce the memory footprint and enhance the computational efficiency of large language models (LLMs) by removing redundant weights while maintaining accuracy. Thanos introduces a block-wise pruning strategy with adaptive masks that dynamically adjust to weight importance, enabling flexible sparsity patterns and structured formats, such as $n:m$ sparsity, optimized for hardware acceleration. Experimental evaluations demonstrate that Thanos achieves state-of-the-art performance in structured pruning and outperforms existing methods in unstructured pruning. By providing an efficient and adaptable approach to model compression, Thanos offers a practical solution for deploying large models in resource-constrained environments.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。