VLMs在哪些隐私属性上达成一致并预测良好？
Which private attributes do VLMs agree on and predict well?
AI总结：
本文研究了VLMs在隐私属性识别上的表现，发现其在某些属性上与人类标注一致且预测准确，可辅助大规模图像数据集的隐私标注。
摘要：
视觉语言模型（VLMs）常用于图像中视觉属性的零样本检测。我们对开源VLMs进行零样本评估，以评估其在隐私属性识别中的表现。我们识别出VLMs在哪些属性上表现出强的跨标注者一致性，并讨论人类和VLM标注之间的不一致情况。我们的结果表明，当与人类标注对比时，VLMs更倾向于预测隐私属性的存在。此外，我们发现当VLMs之间存在高跨标注者一致性时，它们可以补充人类标注，通过识别被人类标注者忽略的属性。这突显了VLMs在大规模图像数据集中的隐私标注潜力。
英文摘要：
Visual Language Models (VLMs) are often used for zero-shot detection of visual attributes in the image. We present a zero-shot evaluation of open-source VLMs for privacy-related attribute recognition. We identify the attributes for which VLMs exhibit strong inter-annotator agreement, and discuss the disagreement cases of human and VLM annotations. Our results show that when evaluated against human annotations, VLMs tend to predict the presence of privacy attributes more often than human annotators. In addition to this, we find that in cases of high inter-annotator agreement between VLMs, they can complement human annotation by identifying attributes overlooked by human annotators. This highlights the potential of VLMs to support privacy annotations in large-scale image datasets.
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。
