Problem
Research questions and friction points this paper is trying to address.
frozen CLIP
region explanations
competing class
CCI
overlap
Beyond Mean Foils: Auditing Worst-Foil Specificity in Frozen CLIP Region Explanations
🤖 AI Summary
研究通过Cluster-based Concept Importance方法在frozen CLIP中测试区域选择，发现大量区域对最强竞争类贡献更大，并探索了替代区域的有效性。
📝 Abstract
A region can overlap a target object yet contribute more to another class. We test regions selected by Cluster-based Concept Importance (CCI) in frozen CLIP. Across COCO and VOC with two checkpoints, 41.08-64.78% of regions that pass overlap and mean-contrast checks fail against the strongest competing class. Removing competitors annotated in the image leaves 39.69-63.64% failing. We then test all eight candidate regions per image. An alternative passes the test for 6.25-7.84% of failures on COCO and 27.40-31.15% on VOC. Requiring it to preserve the original target-score drop within $ε= 0.02$ reduces these rates to 0.16-0.98%. Available regions and target-drop tolerance constrain repair; relaxing the tolerance increases repair opportunities.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Cluster-based Concept Importance (CCI)
frozen CLIP
worst-foil specificity
target-score drop
repair opportunities
🔎 Similar Papers
No similar papers found.
