6 papers
What Makes Synthetic Hard Negatives Work in Vision-Language Pretraining?
Nikos Giakoumoglou, Paschalis Giakoumoglou, Andreas Floros +2
Synthetic hard negatives generated in the representation space have proven effective for unimodal self-supervised learning, but transferring this idea to vision-language pretrainin…
Harmful Content Generation in Text-to-Image Models: Capabilities and Moderation Limitations
Paschalis Giakoumoglou, Manos Schinas, Symeon Papadopoulos
Text-to-image generative models can produce highly realistic imagery but also raise concerns about harmful misuse. While safety mechanisms exist, systematic evaluations of their ef…
Three Necessary Principles for Self-Supervised Visual Representation Learning
Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki
We argue that learning visual representations without labels requires a training signal jointly complete across three non-overlapping objectives: semantic invariance across augment…
DiffusionPrint: Learning Generative Fingerprints for Diffusion-Based Inpainting Localization
Paschalis Giakoumoglou, Symeon Papadopoulos
Modern diffusion-based inpainting models pose significant challenges for image forgery localization (IFL), as their full regeneration pipelines reconstruct the entire image via a l…
TGIF2: Extended Text-Guided Inpainting Forgery Dataset & Benchmark
Hannes Mareen, Dimitrios Karageorgiou, Paschalis Giakoumoglou +3
Generative AI has made text-guided inpainting a powerful image editing tool, but at the same time a growing challenge for media forensics. Existing benchmarks, including our text-g…
SAGI: Semantically Aligned and Uncertainty Guided AI Image Inpainting
Paschalis Giakoumoglou, Dimitrios Karageorgiou, Symeon Papadopoulos +1
Recent advancements in generative AI have made text-guided image inpainting - adding, removing, or altering image regions using textual prompts - widely accessible. However, genera…