Two papers accepted at ICCV 2025

We have two papers accepted at ICCV 2025!

TikZero: Zero-Shot Text-Guided Graphics Program Synthesis, by Jonas Belouadi, Eddy Ilg, Margret Keuper, Hideki Tanaka, Masao Utiyama, Raj Dabre, Steffen Eger, Simone Paolo Ponzetto

Are you interested in scientific figure generation? TikZero, generates scientific figures from text as high-level, human-interpretable, and editable graphics programs. 

AIM: Amending Inherent Interpretability via Self-Supervised Masking, by Eyad Alshami, Shashank Agnihotri, Bernt Schiele and Margret Keuper

Our model AIM improves the interpretability of image classification models by learning to mask out irrelevant features in a self-supervised way.