PhD Researcher · Computer Vision · Semantic Future Prediction · World Models · Generative Modeling
Contact Email: e.karypidis@athenarc.gr
I build systems that predict the future — specifically, how visual scenes will evolve over time. My PhD focuses on semantic future prediction and latent-space world models, using transformers, self-supervised learning, and vision foundation models. I am a PhD Candidate at NTUA and fellow of Archimedes Research Unit / Athena RC.
- Semantic future prediction & world models — forecasting world evolution in learned latent spaces
- Vision foundation models — leveraging and extending large-scale pretrained visual representations
- Generative models × self-supervised learning — bridging generation and representation learning
- Machine unlearning — selective knowledge removal in deep networks
| Paper | Venue | |
|---|---|---|
| 🔥 | DINO-Foresight: Looking into the Future with DINO | NeurIPS 2025 |
| 🔥 | ReDi: Boosting Generative Image Modeling via Joint Image-Feature Synthesis | NeurIPS 2025 |
| 🔥 | FUTURIST: Advancing Semantic Future Prediction through Multimodal Visual Sequence Transformers | CVPR 2025 |