无需训练即可提升视觉模型推理稳定性与准确率
StableTTA: Improving Vision Model Performance by Training-free Test-Time Adaptation Methods
- 基于推理时语义一致性，设计方差感知的对数几率融合策略
- 在ImageNet-1K上71个模型中均实现准确率提升，稳定改进
- 支持单次前向传播，适用于视频、机器人等连续输入场景
集成方法虽能提升预测性能，但常伴随高内存和计算开销。我们发现非线性投影与投票操作引发聚合不稳定性。为解决效率与一致性问题，提出无需训练的Test-Time Adaptation方法StableTTA，含两种变体。StableTTA-I针对相干批次推理场景（如连拍照片、视频流、机器人感知、工业检测），通过方差感知对数几率聚合显著提升预测一致性和准确率。StableTTA-II采用特征级裁剪，仅需一次前向传播即可完成高效对数几率聚合。在ImageNet-1K上对71个模型的实验表明，StableTTA-I在相干批次推理下持续提升准确率，StableTTA-II则以极低计算开销实现轻量级、架构无关的性能增益。结果表明，推理时的语义连贯性与聚合稳定性是优化实际测试时自适应系统的重要视角。
原文摘要 · Abstract (English)
Ensemble methods improve predictive performance but often incur high memory and computational costs. We identify an aggregation instability induced by nonlinear projection and voting operations. To address both efficiency challenges and this inconsistency, we propose StableTTA, a training-free test-time adaptation method with two variants. StableTTA-I targets coherent-batch inference settings, where temporally or semantically adjacent observations are likely to belong to the same class. Examples include burst photography, video streams, robotics perception, and industrial inspection. Under coherent-batch inference, StableTTA-I substantially improves prediction consistency and accuracy through variance-aware logit aggregation. StableTTA-II establishes feature-level cropping, enabling efficient logit aggregation with a single forward pass on a single model backbone. Experiments on ImageNet-1K across 71 models demonstrate that StableTTA-I consistently improves prediction accuracy under coherent-batch inference, while StableTTA-II provides lightweight and architecture-agnostic accuracy improvements with minimal computational overhead. These results suggest that inference-time semantic coherence and aggregation stability provide useful perspectives for improving practical test-time adaptation systems.
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。