Jerome Fink

Jerome Fink is a researcher in the Faculty of Computer Science, focusing on interdisciplinary applications of machine learning and accessibility technologies. His work bridges computer science, linguistics, and medical imaging, with a strong emphasis on sign language recognition and deep learning methodologies.

Key research interests include:

• Sign Language Recognition: Transition modeling, tokenization, and data augmentation techniques using recurrent neural networks and transformers.
• Contrastive Learning: Applications in static sign language recognition and Crohn’s disease detection via endoscopic data.
• Software Security: Automated risk assessment frameworks for shell-based attacks leveraging large language models.

His recent research outputs (2024–2025) demonstrate a focus on hybrid modeling approaches, cross-domain data integration, and accessibility-driven AI. Fink received the Best NIER Paper Award at SCAM 2020 for innovative contributions to source code analysis. Collaborative activities include organizing international seminars on sign languages and participating in AI conferences (IJCAI 2023). Supervised theses and student projects highlight his role in mentoring next-generation researchers in data science and deep learning.