MoST用稀疏矩阵高效微调3D点云模型，不增加推理开销且性能领先。
MoST: Efficient Monarch Sparse Tuning for 3D Representation Learning
- 将密集更新矩阵重参数化为稀疏的点云专用结构矩阵
- 在ScanObjectNN上达97.5%准确率，ModelNet40达96.2%
- 兼容多种3D骨干网络，适合资源受限场景下的高效微调
我们提出面向3D表示学习的首个基于重参数化的参数高效微调方法MoST。与现有适配器或提示微调方法不同，MoST不引入额外推理开销，兼容多种3D表示学习骨干网络。核心是提出一类用于3D点云的新结构矩阵——Point Monarch，能捕捉不规则点的局部几何特征并具备高表达能力。MoST将密集更新权重矩阵重参数化为稀疏的Point Monarch矩阵，在显著减少参数量的同时保持强性能。实验表明，该方法简单有效、高度通用，能捕捉点云局部特征，在多个基准上取得最先进结果：如ScanObjectNN (PB_50_RS) 上达到97.5%准确率，ModelNet40分类任务中达96.2%。同时可与低秩、Kronecker等矩阵分解技术结合，进一步压缩参数。
原文摘要 · Abstract (English)
We introduce Monarch Sparse Tuning (MoST), the first reparameterization-based parameter-efficient fine-tuning (PEFT) method tailored for 3D representation learning. Unlike existing adapter-based and prompt-tuning 3D PEFT methods, MoST introduces no additional inference overhead and is compatible with many 3D representation learning backbones. At its core, we present a new family of structured matrices for 3D point clouds, Point Monarch, which can capture local geometric features of irregular points while offering high expressiveness. MoST reparameterizes the dense update weight matrices as our sparse Point Monarch matrices, significantly reducing parameters while retaining strong performance. Experiments on various backbones show that MoST is simple, effective, and highly generalizable. It captures local features in point clouds, achieving state-of-the-art results on multiple benchmarks, e.g., 97.5% acc. on ScanObjectNN (PB_50_RS) and 96.2% on ModelNet40 classification, while it can also combine with other matrix decompositions (e.g., Low-rank, Kronecker) to further reduce parameters.
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。