Postdoctoral Position: Data-efficient Detection of AI-generated Images and Videos Using Multimodal Foundation Models

The project aims to create reliable forensic analysis tools to detect AI-generated images and videos, addressing the challenges posed by generative AI. Candidates will work on using multimodal foundation models and weakly-supervised learning methods, validating their approaches against standard benchmarks. The role requires a PhD in a relevant field, experience in deep learning, and proficiency in Python. The working language may be French or English.