Multimodal Deep Learning and Information Retrieval
Representation alignment, composed image retrieval, and retrieval across historical collections.
My Ph.D. research at the University of Luxembourg focuses on:
- Multimodal representation alignment for cross-modal information retrieval.
- Training-free and zero-shot composed image retrieval with multimodal large language models and diffusion models.
- Cross-modal and cross-lingual retrieval across historical image and text collections.