发现开放词汇检测的置信度受尺度和语义双重干扰，不可靠。
Confidence Scores in Open-Vocabulary Detection Are a Biased Mixture of Scale and Semantics
- 通过控制实验揭示置信度受物体大小和查询语义影响
- 大物体得分偏高，通用查询得分被抑制，偏差显著且普遍
- 提出无需训练的温度校正法，提升小物体召回率19.6%
基于CLIP等基础模型的开放词汇检测器虽能泛化至新类别，但其置信度分数并非可靠的定位概率估计，而是视觉尺度与语义查询特异性共同作用的混合结果。在COCO数据集上对GroundingDINO、OWL-ViT、YOLO-World三款检测器进行控制实验，结合在LVIS（1,203类）上使用GroundingDINO的复现结果表明，余弦相似度s=cos(v,t)存在系统性偏差：尺度偏差（alpha = +0.064，r = 0.579，p = 1.29 × 10^-58）使大物体得分偏高；语义偏差（beta = -0.705，p = 5.23 × 10^-41）抑制通用查询得分。该偏差源于CLIP图像级预训练的固有结构。阈值调整无法消除偏差：最优按尺度调阈后，小物体F1仅提升0.001，大物体提升0.102。一种无参数的温度缩放校正方法在不重新训练的情况下，将小物体Recall@10提升19.6%（p < 0.01），但会轻微牺牲整体排序精度，说明偏差可部分但非完全在推理时修正。该研究揭示了将图像级基础模型应用于区域级检测任务的根本局限。
原文摘要 · Abstract (English)
Foundation models such as CLIP have enabled open-vocabulary object detectors that generalise to novel categories via vision-language similarity. However, the confidence scores these detectors produce are not reliable localization probability estimates: they conflate visual scale and semantic query specificity with the true detection signal. Through controlled experiments on COCO across three foundation-model-based detectors (GroundingDINO, OWL-ViT, YOLO-World), with the scale-bias finding further replicated on LVIS (1,203 categories) using GroundingDINO, we show that s=cos(v,t) is a biased mixture of two effects. Scale bias (alpha = +0.064, r = 0.579, p = 1.29 x 10^-58) systematically inflates scores for large objects. Semantic bias (beta = -0.705, p = 5.23 x 10^-41) suppresses scores for generic queries. Both biases are structurally inevitable from CLIP's image-level pretraining. Threshold adjustment cannot remove them: oracle per-scale thresholding yields Delta F1 = +0.001 for small objects versus +0.102 for large. A parameter-free temperature scaling correction improves small-object Recall@10 by 19.6% (p < 0.01) without retraining. This comes at a modest, measurable cost to pooled-ranking precision, so the bias is partially, not freely, reversible at inference time. These findings reveal a fundamental limitation of adapting image-level foundation models to region-level detection tasks.
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。