SABOKROU MOHAMMAD EBRAHIM

Mohammad Sabokrou

Associate Professor, Department of Artificial Intelligence, School of Computing

New Uzbekistan University, Tashkent, Uzbekistan

Visiting Researcher, Okinawa Institute of Science and Technology (OIST), Japan

Email:  

Personal website: https://sabokrou.github.io/

Google Scholar: Profile

Mohammad Sabokrou is an Associate Professor in the Department of Artificial Intelligence, School of Computing, at New Uzbekistan University, and a Visiting Researcher at the Okinawa Institute of Science and Technology (OIST), Japan.

His research focuses on trustworthy artificial intelligence, machine learning, and computer vision — in particular anomaly and novelty detection, out-of-distribution generalization, AI safety and security, adversarial robustness, continual learning, and reliable vision–language models.

He has held visiting research positions at the University of Technology of Troyes (France), Fudan University (China), and Tokyo University of Science (Japan), and maintains active international collaborations with researchers at leading institutions worldwide, including OIST, the University of Tokyo, the University of Oulu, Stanford University, and Sharif University of Technology, among others across Asia, Europe, and North America.

1. Trustworthy artificial intelligence and machine learning
2. AI safety, security, and adversarial robustness
3. Anomaly, novelty, and out-of-distribution detection
4. Continual and lifelong learning
5. Vision–language and self-supervised learning
6. Reliable and robust computer vision

1. Breaking Boundaries: Robust, Domain-General Anomaly Detection with Vision–Language Models. JSPS KAKENHI Grant-in-Aid for Scientific Research (B), 2026–2030 (¥18,330,000).
2. Investigating the Trustworthiness of Deep Pre-trained and Self-Supervised Models. JSPS KAKENHI Grant-in-Aid for Early-Career Scientists, 2024–2028 (¥4,680,000).
3. AI Safety and Security for Classical AI Models. Institute for Research in Fundamental Sciences (IPM).

Area Chair: ICLR (2025, 2026, 2027), NeurIPS (2026), and BMVC (2026).

Program Committee member and reviewer for major AI, machine-learning, and computer-vision venues, including NeurIPS, ICLR, CVPR, ICCV, ECCV, AAAI, WACV, MICCAI, and ACCV.

Contributor to the organization of international machine-learning events, workshops, and summer schools, including the Machine Learning Summer School (MLSS) and research events at OIST.

1. M. Pourkeshavarz, G. Zhao, Mohammad Sabokrou. Looking Back on Learned Experiences for Class/Task Incremental Learning. International Conference on Learning Representations (ICLR), 2022. [Spotlight]
2. M. Salehi, H. Mirzaei, D. Hendrycks, Y. Li, M. H. Rohban, Mohammad Sabokrou. A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection. Transactions on Machine Learning Research (TMLR), 2022.
3. Mohammad Sabokrou, M. Khalooei, E. Adeli. Self-Supervised Representation Learning via Neighborhood-Relational Encoding. IEEE/CVF International Conference on Computer Vision (ICCV), 2019.
4. Mohammad Sabokrou, M. Khalooei, M. Fathy, E. Adeli. Adversarially Learned One-Class Classifier for Novelty Detection. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2018.
5. M. Nafez, M. Poulaei, N. Vasei, B. Soltani Moakhar, Mohammad Sabokrou, M. H. Rohban. FrameShield: Adversarially Robust Video Anomaly Detection. Advances in Neural Information Processing Systems (NeurIPS), 2025.
6. S. Sharifipour, C. Álvarez Casado, Mohammad Sabokrou, M. Bordallo López. APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction. Advances in Neural Information Processing Systems (NeurIPS), 2025.
7. H. Mirzaei, M. Nafez, J. Habibi, Mohammad Sabokrou, M. H. Rohban. Mitigating Spurious Negative Pairs for Robust Industrial Anomaly Detection. International Conference on Learning Representations (ICLR), 2025.
8. H. Mirzaei, A. Ansari, B. Dibaei Nia, M. Nafez, M. Madadi, S. Rezaee, Z. S. Taghavi, A. Maleki, K. Shamsaie, M. Hajialilue, J. Habibi, Mohammad Sabokrou, M. H. Rohban. Scanning Trojaned Models Using Out-of-Distribution Samples. Advances in Neural Information Processing Systems (NeurIPS), 2024.
9. H. Mirzaei, M. Nafez, M. Jafari, M. B. Soltani, M. Azizmalayeri, J. Habibi, Mohammad Sabokrou, M. H. Rohban. Universal Novelty Detection Through Adaptive Contrastive Learning. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
10. M. Pourkeshavarz, Mohammad Sabokrou, A. Rasouli. Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Driving. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024.

A complete and up-to-date list of publications is available on his personal website and Google Scholar profile.