Tejaswini Medi

Researcher

University of Mannheim

B6, 28–29, Room C 001, 68159, Mannheim

Email: tejaswini.medimail-uni-mannheim.de

Webpage: tejaswinimedi.github.io

Linkedin Twitter Bluesky Social Google Scholar

Research Interests

Robustness and Fairness of Deep Learning Models
Image and 3D Generation
Representation Learning

Publications

Tejaswini Medi, Julia Grabinski, Margret Keuper, “Towards Class-wise Robustness Analysis” accepted at OOD Generalization in Computer Vision Workshop, ICCV 2023.Tejaswini Medi, Jawad Tayyub, Muhammad Sarmad, Frank Lindseth, Margret Keuper, “FullFormer: Generating Shapes Inside Shapes”, accepted at DAGM German Conference on Pattern Recognition, Heidelberg, 2023. Tejaswini Medi, Steffen Jung, Margret Keuper, " FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Training”, accepted at Proceedings of the Winter Conference on Applications of Computer Vision (WACV), 2025

Teaching

Teaching Assistant

Generative Computer Vision Models ( FSS 2024, FSS 2025)
Machine Learning ( HWS 2024)
Seminar Organization (AI Safety & Robustness) ( FSS 2025)
Seminar Supervision ( HWS 2024 – FSS 2025)

Community Participation

Active Reviewer for TPAMI, CVPR and NeurIPS Conferences