Glimpse from the Lab: The Fingerprint Project
Data-free LoRA indexing and dynamic adaptation for open-vocabulary semantic segmentation under real-world domain shifts
Mehdi is an MSc student in Computer Science at Sapienza University of Rome. His interests include model merging, domain adaptation, and vision models, and he is particularly motivated by multimodal AI and real-world applications. Outside research, he enjoys spending time with friends, listening to house music, and watching historical documentaries.
Data-free LoRA indexing and dynamic adaptation for open-vocabulary semantic segmentation under real-world domain shifts