We perform extensive experiments to confirm the effectiveness of AnyChange for zero-shot change detection.

To alleviate the above issues, we propose a simple yet effective framework for producing H igh Q uality black-box hard-label A dversarial Attack, named HQA-Attack . The overview of HQA-Attack is shown in Figure 1. By "high quality", it means that the HQA-Attack method can generate

Our work naturally extends Word Mover's Distance (WMD) [

In this supplementary, we provide additional details, results and visualizations.

Althoughintuitive,sucha nativelabelassignment strategycannot revealtheunderlying semantic similarity between aquery anditspositivesandnegatives,andimpairs performance, since some negatives are semantically similar to the query or even share the same semantic class as the query.

Object detection has achieved promising success, but requires large-scale fullyannotated data, which is time-consuming and labor-extensive. Therefore, we consider object detection with mixedsupervision, which learns novelobject categories using weak annotations with thehelpoffullannotations ofexistingbase objectcategories.