Scholar
Dheeraj Varghese
Google Scholar ID: f5Yht7IAAAAJ
University of Amsterdam
Citations & Impact
Citations
0
H-index
0
i10-index
0
Publications
3
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- Published 'Vision-and-Language Training Helps Deploy Taxonomic Knowledge but Does Not Fundamentally Alter It' (NeurIPS, 2025); released 'NeoBabel: A Multilingual Open Tower for Visual Generation' (arXiv preprint arXiv:2507.06137, 2025); served as a reviewer for ICCV LIMIT Workshop 2025; co-organized a hackathon for SAGE-MLU 2025.
Research Experience
- Worked on generalist multimodal foundation models as part of the Horizon Europe ELLIOT project; collaborated with Mohammad M. Derakhshani on multilingual image generation; explored curriculum learning in vision-language models under the supervision of Yuki Asano.
Education
- PhD candidate at VIS Lab, supervised by Cees Snoek.
Background
- Research interests include designing unified architectures to combine modalities into a shared space, aiming for models that generalize well, adapt efficiently, and assist meaningfully across a wide range of tasks. At the core, an applied engineer with enthusiasm for recreating intelligence as a tool to make tasks easier for the human user.
Miscellany
- Will be a Teaching Assistant for Natural Language Processing at VU starting March 21, 2024.
