Publications

2026

• Coevolving Representations in Joint Image-Feature Diffusion

  T. Kouzelis, S. Gidaris, N. Komodakis · arXiv:2604.17492

• Disentangling Local and Global Semantics in Diffusion Models for Image Editing

  M. Plitsis, T. Kouzelis, P. Koromilas, V. Katsouros, M. A. Nicolaou, et al. · International Journal of Computer Vision

2025

• Boosting Generative Image Modeling via Joint Image-Feature Synthesis

  T. Kouzelis, E. Karypidis, I. Kakogeorgiou, S. Gidaris, N. Komodakis · NeurIPS 2025 Spotlight

• EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling

  T. Kouzelis, I. Kakogeorgiou, S. Gidaris, N. Komodakis · ICML 2025

2024

• Enabling Local Editing in Diffusion Models by Joint and Individual Component Analysis

  T. Kouzelis, M. Plitsis, M. A. Nicolaou, Y. Panagakis · BMVC 2024

• Investigating Personalization Methods in Text to Music Generation

  M. Plitsis*, T. Kouzelis*, G. Paraskevopoulos, V. Katsouros, Y. Panagakis · ICASSP 2024

2023

• Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems: A Case Study for Modern Greek

  G. Paraskevopoulos, T. Kouzelis, G. Rouvalis, A. Katsamanis, V. Katsouros, et al. · IEEE/ACM Transactions on Audio, Speech, and Language Processing

• Weakly-Supervised Automated Audio Captioning via Text-Only Training

  T. Kouzelis, V. Katsouros · DCASE 2023

• Weakly-Supervised Forced Alignment of Disfluent Speech Using Phoneme-Level Modeling

  T. Kouzelis, G. Paraskevopoulos, A. Katsamanis, V. Katsouros · INTERSPEECH 2023

• Efficient Audio Captioning Transformer with Patchout and Text Guidance

  T. Kouzelis, G. Bastas, A. Katsamanis, A. Potamianos · DCASE 2023 Challenge