利用状态空间模型特性，在测试时动态优化参数以提升鲁棒性。
TRUST: Test-Time Refinement using Uncertainty-Guided SSM Traverses
- 通过多种遍历路径生成图像的多重因果视角
- 基于预测伪标签更新Mamba参数并平均融合结果
- 首个针对SSM架构设计的测试时自适应方法
状态空间模型（SSMs）已成为视觉变换器（ViTs）的高效替代方案，其中VMamba是专为视觉任务设计的开创性架构。然而，其在分布外场景下的泛化性能显著下降。为此，我们提出TRUST（Test-Time Refinement using Uncertainty-Guided SSM Traverses），一种新颖的测试时自适应（TTA）方法，利用多样化的遍历排列生成输入图像的多个因果视角。模型预测作为伪标签，引导Mamba特有参数的更新，适应后的权重通过平均整合各遍历扫描中学习到的信息。总体而言，TRUST是首个明确利用SSM独特架构特性进行自适应的方法。在七个基准上的实验表明，TRUST能持续提升鲁棒性，并优于现有TTA方法。
原文摘要 · Abstract (English)
State Space Models (SSMs) have emerged as efficient alternatives to Vision Transformers (ViTs), with VMamba standing out as a pioneering architecture designed for vision tasks. However, their generalization performance degrades significantly under distribution shifts. To address this limitation, we propose TRUST (Test-Time Refinement using Uncertainty-Guided SSM Traverses), a novel test-time adaptation (TTA) method that leverages diverse traversal permutations to generate multiple causal perspectives of the input image. Model predictions serve as pseudo-labels to guide updates of the Mamba-specific parameters, and the adapted weights are averaged to integrate the learned information across traversal scans. Altogether, TRUST is the first approach that explicitly leverages the unique architectural properties of SSMs for adaptation. Experiments on seven benchmarks show that TRUST consistently improves robustness and outperforms existing TTA methods.
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。