Publications
A. Alexos, R. Peri, S. M. Jayanthi, M. Cekic, S. Vishnubhotla, K. J. Han, S. Ronanki, "Defending Speech-enabled LLMs Against Adversarial Jailbreak Threats" In Proc. Interspeech 2025 (pp. 2048-2052), Rotterdam, Netherlands, Aug 2025.
L. Goncalves, P. Mathur, C. Lavania, M. Cekic, M. Federico, K. J. Han, "PEAVS: Perceptual Evaluation of Audio-Visual Synchrony Grounded in Viewers' Opinion Scores", The 18th European Conference on Computer Vision (ECCV), Milano, Italy, Sep 2024.
M. Cekic, C. Bakiskan, U. Madhow, "Neuro-Inspired Deep Neural Networks with Sparse, Strong Activations", IEEE International Conference in Image Processing (ICIP), Bordeaux, France, Oct 2022.
M. Cekic, C. Bakiskan, and U. Madhow. "Towards robust, interpretable neural networks via Hebbian/anti-Hebbian learning: A software framework for training with feature-based costs.", Software Impacts (2022).
M. Cekic, R. Li, Z. Chen, Y. Yang, A. Stolcke, U. Madhow, "Self-Supervised Speaker Recognition Training Using Human-Machine Dialogues", IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, May 2022.
M. Cekic, S. Gopalakrishnan, U. Madhow, "Wireless Fingerprinting via Deep Learning: The Impact of Confounding Factors", IEEE Asilomar Conference on Signals, Systems, and Computers, Nov. 2021
C. Bakiskan, M. Cekic, A. D. Sezer, U. Madhow, "A Neuro-Inspired Autoencoding Defense Against Adversarial Perturbations", IEEE International Conference on Image Processing (ICIP), Anchorage, Sept. 2021
C. Bakiskan, S. Gopalakrishnan, M. Cekic, U. Madhow, R. Pedarsani, "Polarizing Front Ends For Robust CNNs", IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, May 2020
S. Gopalakrishnan, M. Cekic, U. Madhow, "Robust Wireless Fingerprinting via Complex-Valued Neural Networks", IEEE Global Communications Conference (Globecom), Hawaii, Dec. 2019.
Workshop Proceedings
M. Cekic, C. Bakiskan, U. Madhow, "Layerwise Hebbian/anti-Hebbian (HaH) Learning In Deep Networks: A Neuro-inspired Approach To Robustness", International Conference on Machine Learning (ICML) 2022 Workshop – New Frontiers in Adversarial Machine Learning, Baltimore, USA, Jul 2022.
C. Bakiskan, M. Cekic, U. Madhow, "Early Layers Are More Important For Adversarial Robustness", International Conference on Machine Learning (ICML) 2022 Workshop – New Frontiers in Adversarial Machine Learning, Baltimore, USA, Jul 2022.
C. Bakiskan, M. Cekic, A. D. Sezer, U. Madhow, "Sparse Coding Frontend For Robust Neural Networks", International Conference on Learning Representations (ICLR), Workshop on Security and Safety in Machine Learning Systems, May 2021.
Dissertation
M. Cekic, "Robust Learning Techniques for Deep Neural Networks", Doctoral dissertation, UC Santa Barbara, 2022.
Preprints
S. Gopalakrishnan, Z. Marzi, M. Cekic, U. Madhow, R. Pedarsani, "Robust Adversarial Learning via Sparsifying Front Ends", Preprint, Arxiv.