PrivLEX: Detecting legal concepts in images through Vision-Language Models

2026-01-14

AI Summary

This study addresses the challenge of identifying legally defined personal data concepts in images to enable interpretable privacy classification. To this end, the authors propose a novel zero-shot approach that requires no explicit concept labels, integrating vision-language models with an unsupervised concept bottleneck model. This method is the first to align personal data concepts derived from legal texts with image-based privacy classification. It not only accurately detects legally relevant sensitive concepts in images but also demonstrates strong human-perceived consistency of these concepts, as validated through user studies. The work establishes a new paradigm for interpretable and legally compliant privacy protection by bridging the gap between regulatory definitions and visual data analysis.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Bias, Fairness & Privacy
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

We present PrivLEX, a novel image privacy classifier that grounds its decisions in legally defined personal data concepts. PrivLEX is the first interpretable privacy classifier aligned with legal concepts that leverages the recognition capabilities of Vision-Language Models (VLMs). PrivLEX relies on zero-shot VLM concept detection to provide interpretable classification through a label-free Concept Bottleneck Model, without requiring explicit concept labels during training. We demonstrate PrivLEX's ability to identify personal data concepts that are present in images. We further analyse the sensitivity of such concepts as perceived by human annotators of image privacy datasets.

Problem

• privacy classification
• legal concepts
• personal data
• Vision-Language Models
• image privacy

Innovation

• Vision-Language Models
• zero-shot concept detection
• Concept Bottleneck Model
• interpretable privacy classification
• legal concepts