Which private attributes do VLMs agree on and predict well?
| Type of publication: | Conference paper |
| Citation: | Hrynenko_ICASSP_2026 |
| Booktitle: | Proceedings of the 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
| Year: | 2026 |
| Abstract: | 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. |
| Main Research Program: | AI for Everyone |
| Keywords: | attributes recognition, privacy, VLM |
| Authors: | |
| Added by: | [UNK] |
| Total mark: | 0 |
|
Attachments
|
|
|
|
|
Notes
|
|
|
|
|
