Research

We specialize in multimodal modeling approaches that address complex problems spanning various domains. Our work encompasses the development of novel machine learning algorithms that can effectively handle and integrate different data modalities while considering various levels of scale and abstraction. This includes advancing methods in computer vision, multimodal information retrieval, natural language processing, and human-centered AI applications.

Our research has strong interdisciplinary connections, collaborating with colleagues in digital humanities, education, psychology/psychiatry, neurosciences, medicine, and environmental modeling. We believe in the power of AI to bridge disciplines and create innovative solutions for societal challenges.

Current Research Areas

• Multimodal Data Analysis: Developing methods to analyse and understand multimodal data (visual, audiovisual, textual, and sensory information)
• Multimodal Information Retrieval: Multimodal search and recommendation systems
• Computer Vision: Advancing image and video analysis methods for various applications
• Vision-Language Models: Advancing the integration of visual and textual understanding
• Educational Technology: AI-assisted learning systems and educational content analysis
• Human-Centered AI: Developing AI systems that support human decision-making and learning