PAF：面向开放集半监督学习的扰动感知过滤
PAF: Perturbation-Aware Filtering for Open-Set Semi-Supervised Learning
把「语义保持扰动下表征是否稳定」当作区分 ID/OOD 的表示级信号，可直接迁移到任何需要从无标注池中筛掉分布外样本的视觉训练流程（如开放世界分类、半监督检测的伪标签过滤）。
原文依据 PAF identifies OOD samples by measuring the representation instability under semantic-preserving perturbations.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论开放集半监督学习（OSSL）在利用无标注数据方面已取得显著进展，但现有方法大多忽略了分布内（ID）与分布外（OOD）样本对语义保持扰动的敏感性差异，导致模型性能次优。为解决这一局限，我们提出扰动感知过滤（PAF），它利用 ID 与 OOD 样本在扰动下的行为差异，并将其扩展为表示层面的信号以实现可靠的 OOD 过滤。具体而言，PAF 通过度量语义保持扰动下的表征不稳定性来识别 OOD 样本。随后我们将 PAF 集成到一个精心设计的两阶段训练框架中，使模型在第一阶段能够利用丰富的无标注数据，并在第二阶段借助有限的标签逐步适应开放集设定。在广泛使用的 OSSL 基准上的大量实验结果表明，我们提出的 PAF 方法相比最优（SOTA）的 OSSL 方法取得了更优的性能。我们的代码已发布于 https://github.com/njustkmg/CVPR26-PAF。
英文原文摘要 1159 字符
Open-set semi-supervised learning (OSSL) has achieved notable progress in exploiting unlabeled data, yet most existing methods overlook the distinct sensitivities of in-distribution (ID) and out-of-distribution (OOD) samples to semantic-preserving perturbations, resulting in suboptimal model performance. To address this limitation, we propose Perturbation-Aware Filtering (PAF), which leverages the behavioral difference between ID and OOD samples under perturbations and extends it into a representation-level signal for reliable OOD filtering. Specifically, PAF identifies OOD samples by measuring the representation instability under semantic-preserving perturbations. We then integrate PAF into a carefully designed two-stage training framework, allowing the model to exploit abundant unlabeled data in the first stage and gradually adapt to the open-set setting with limited labels in the second stage. Extensive experimental results on widely-used OSSL benchmarks demonstrate that our proposed PAF approach achieves superior performance compared to state-of-the-art (SOTA) OSSL methods. Our code is available at https://github.com/njustkmg/CVPR26-PAF.