通过清除受域偏移影响的点云令牌，实现无需反向传播的快速自适应分类
Purge-Gate: Backpropagation-Free Test-Time Adaptation for Point Clouds Classification via Token Purging
- 在注意力层前清除受域偏移影响的点云令牌，避免迭代更新
- 在多个测试数据集上比现有无反向传播方法平均提升10.3%准确率
- 速度比基线快12.4倍，内存效率高5.5倍，适合实际部署
测试时自适应（TTA）对于缓解3D点云分类中分布偏移导致的性能下降至关重要。本文提出一种名为令牌清除（Token Purging, PG）的新方法，这是一种无需反向传播的全新策略，在令牌进入注意力层前清除受域偏移显著影响的项。与现有方法不同，PG在令牌层面操作，确保无需迭代更新即可实现鲁棒自适应。我们提出两种变体：利用源数据统计信息的PG-SP，以及完全无源的PG-SF，后者依赖CLS令牌驱动自适应。在ModelNet40-C、ShapeNet-C和ScanObjectNN-C上的大量实验表明，PG-SP比当前最先进的无反向传播方法平均提升10.3%准确率，而PG-SF为无源自适应设定了新基准。此外，PG比基线速度快12.4倍，内存效率高5.5倍，适用于真实场景部署。代码已开源。
原文摘要 · Abstract (English)
Test-time adaptation (TTA) is crucial for mitigating performance degradation caused by distribution shifts in 3D point cloud classification. In this work, we introduce Token Purging (PG), a novel backpropagation-free approach that removes tokens highly affected by domain shifts before they reach attention layers. Unlike existing TTA methods, PG operates at the token level, ensuring robust adaptation without iterative updates. We propose two variants: PG-SP, which leverages source statistics, and PG-SF, a fully source-free version relying on CLS-token-driven adaptation. Extensive evaluations on ModelNet40-C, ShapeNet-C, and ScanObjectNN-C demonstrate that PG-SP achieves an average of +10.3\% higher accuracy than state-of-the-art backpropagation-free methods, while PG-SF sets new benchmarks for source-free adaptation. Moreover, PG is 12.4 times faster and 5.5 times more memory efficient than our baseline, making it suitable for real-world deployment. Code is available at \hyperlink{https://github.com/MosyMosy/Purge-Gate}{https://github.com/MosyMosy/Purge-Gate}
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。