arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.17818cs.CV
收藏

基于证据的未知拒绝用于高置信度近似未知物

Evidence-Guided Unknown Rejection for High-Confidence Near-Known Unknowns

Xi Chen 1(陈曦 1);Yingjun Xiao 2(肖英俊 2);Gang Fang 3(方刚 3)

Xi Chen, Yingjun Xiao, Gang Fang

AI总结:

本文提出EGUR-A方法,通过改变决策方式从判断样本得分是否足够高到判断预测已知类别是否有足够证据接受样本,从而减少高置信度的误判接受。

摘要:

开放集识别系统面临一个被忽视的失败模式:高置信度的近似未知物,这些样本位于已知标签集之外,但足够接近已知类别,使得闭合集分类器以高置信度接受它们。我们证明这种失败在标量阈值方法中普遍存在,包括最近的后处理检测器,并且更强的编码器可能放大而非消除风险。我们提出EGUR-A,将决策从『这个样本的得分是否足够高?』转变为『这个预测的已知类别是否有足够的证据来接受这个样本?』EGUR-A结合类别条件的局部接受证据与全局残差证据,并从已知样本统计中选择其相对权重,而无需未知验证数据。在CUB、FGVC-Aircraft和ImageNet-hard上,EGUR-A显著减少了在匹配已知拒绝操作点处的高置信度误判接受。结果不是更强的阈值,而是不同的问题:已知类别是否有权接受样本。

英文摘要:

Open-set recognition systems face a neglected failure mode: high-confidence near-known unknowns, which lie outside the known label set but are close enough to known classes that a closed-set classifier accepts them with high confidence. We show that this failure is widespread across scalar-threshold methods, including recent post-hoc detectors, and that stronger encoders can amplify rather than remove the risk. We propose EGUR-A, which changes the decision from ``is this sample's score high enough?'' to ``does this predicted known class have sufficient evidence to accept this sample?'' EGUR-A combines class-conditional local acceptance evidence with global residual evidence, and selects their relative weight from known-sample statistics without unknown validation data. Across CUB, FGVC-Aircraft, and ImageNet-hard, EGUR-A substantially reduces high-confidence false known acceptance at matched known-rejection operating points. The result is not a stronger threshold; it is a different question: whether a known class is entitled to accept a sample.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑