Stay updated with the latest research breakthroughs, publications, achievements, and milestones from the Secure Prediction, Edge AI, and LLM Lab.
๐ Received NSF NAIRR Pilot Resource Allocation Grant! Thanks NSF! [Link]
๐ SPEED LAB Organized a Special Session on Robustness and Security of Large Language Models (ROSE-LLM 2025) at the 24th IEEE International Conference on Machine Learning and Applications (ICMLA 2025)!
๐ Three papers are presented at ICMLA 2025!
๐ Our paper, titled "Quantifying Robustness and Sustainability Trade-off in Federated Adversarial Learning for Cyber-Physical Systems" is accepted at IEEE Transactions on Sustainable Computing!
๐Dr. Imteaj is Listed as World top 2% of scientists in the annual StanfordโElsevier ranking!
๐ Our paper, titled "Pose to Protect: Federated Skeleton-Based Anomaly Detection for Privacy-Conscious Video Surveillance" is presented at ICCV 2025 in Honolulu, Hawaii, one of the premier conferences in computer vision!
๐ Three papers are accepted in ICMLA'25!
๐ Dr. Imteaj joined as a Tenure-track Assistant Professor in the Department of Electrical Engineering and Computer Science and I-SENSE Fellow at Florida Atlantic University.
Paper titled 'Towards Trustworthy Autonomous Vehicles with Vision-Language Models Under Targeted and Untargeted Adversarial Attacks' accepted at CVPR 2025 Fair, Data-efficient, and Trusted Computer Vision Workshop.
Our paper 'SLADE: Shielding against Dual Exploits in Large Vision-Language Models' got accepted at CVPR'25.
Released a survey on Distributed and Multimodal LLMs led by FIU.
Paper accepted at AAAI 2025 Spring Symposium.
Published co-first authored paper in IEEE Transactions on AI.
Dr. Ahmed Imteaj has been nominated for the SIU Early Career Faculty Excellence Award.
Md Zarif Hossain, a member of our research team, presented two papers at IEEE BigDataโ24 and ICMLAโ24.
Dr. Ahmed Imteaj has been awarded the ORAU Research Innovation Partnerships Grant as the Lead PI. We thank ORAU for their support!
Our paper titled 'Securing Vision-Language Models with a Robust Encoder against Jailbreak and Adversarial Attacks' has been published at IEEE BigData'24. This research focuses on developing a robust encoder to secure vision-language models against jailbreak and adversarial attacks, ensuring the integrity and reliability of AI systems.
Dr. Ahmed Imteaj, Director of the SPEED Lab has been listed in the prestigious World's Top 2% Scientist list, as recognized by Stanford University. This recognition highlights his outstanding contributions to research with significant impact in 2023. The list, compiled by Stanford University and published by Elsevier, ranks scientists across various fields based on standardized citation metrics, such as citation counts, h-index, and co-authorship metrics.
Md Zarif Hossain, a member of our research team, presented our paper titled 'FLAMINGO: Adaptive and Resilient Federated Meta-Learning against Adversarial Attacks' at the ICDCS workshop held in New Jersey.
We are excited to welcome Oleksandr, our newest MS student, to the SPEED Lab! Oleksandr will be working on the intersection of federated learning and VLM. Welcome to SPEED Lab, Oleksandr!
We are excited to welcome Awal Ahmed Fime, our newest PhD student, to the SPEED Lab! Awal's will be working on improving perception and decision-making of Connected Autonomous Vehicles. Welcome to SPEED Lab, Awal!
April 2024 - ๐ Dr. Ahmed Imteaj, a distinguished faculty member at the School of Computing, has been awarded the prestigious National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Research Initiation Initiative (CRII) grant of amount $167,000 as a sole PI. This accolade includes a substantial sum of $167,500, provided by the NSF's Division of Information and Intelligent Systems. Thanks to NSF!
April 2024 - ๐ Congrats to SPEED Lab Director, Dr. Imteaj for receiving the Outstanding Teacher of the year award from the School of Computing at Southern Illinois University!
March 2024 - ๐ Dr. Ahmed Imteajโs collaborative project with FIU has been awarded a grant from US Department of Homeland Security (DHS) under the Criminal Investigations and Network Analysis Center (CINA) program. The project is scheduled for a duration of one year, with an approved funding amount of around $100,000. Thanks to DHS!
April 2024 Two papers from the SPEED Lab is accepted at the prestigious 44th IEEE International Conference on Distributed Computing Systems (ICDCS'24) conference. Congrats to PhD student, Zarif and Dr. Imteaj!
March 2024- Four papers from the SPEED Lab is accepted at the IEEE COMPSAC'24 conference.
December 2023 SPEED lab member, Md Zarif Hossain, has been awarded with a prestigious Doctoral fellowship in recognition of his exceptional academic achievements and potential in research. Zarifโs unwavering commitment to advancing the boundaries of our understanding through rigorous research has distinguished him among his peers, making this accolade a well-deserved acknowledgment of his hard work and dedication. Congratulations, Zarif!
November 23 SPEED Lab member, Saika Zaman won the Second Prize for her paper on Trustworthy Agent-Driven Particle Swarm Optimization for Communication-Efficient Federated Learning for Communication-Efficient Federated Learning at the 2023 CMD-IT/ACM Richard Tapia Celebration of Diversity in Computing Conference in Dallas, Texas, and she has been awarded the prestigious Tapia scholarship. Congratulations, Saika!
Dr. Imteaj will present the paper at Vancouver!
A demonstration of our work can be found at https://youtu.be/5gD1Te1Fgnw!
PhD Student Md Zarif Hossain presented the paper in the conference!
Dr. Imteaj presented the paper in Washington DC!
SPEED Lab member, Gireesh Nadh Mekala successly completed his MS project on Federated Learning-driven Real-time Traffic Analysis and Forecasting using Social Media Data under the supervision of Dr. Ahmed Imteaj. Congrats, Gireesh!
Our paper on Federated Learning and Blockchain in collaboration with FIU is published at IEEE Internet of Things Journal (Impact Factor: 10.238)!
Our paper, Assessing Wearable Human Activity Recognition Systems Against Data Poisoning Attacks in Differentially-Private Federated Learning, has been accepted to IEEE SmartComp'23! SPEED Lab PhD student Md. Zarif Hossain presented the paper at Nashville, TN
SPEED Lab PhD Student, Md Zarif Hossain has been awarded with prestigious NSF Student Travel Grant to attend 2023 ACM MobiHoc. Congratulations, Zarif!
SPEED Lab former undergraduate student researcher, Thomas Boyd joined at a startup as a ML Engineer. Congrats, Boyd!
SPEED Lab former undergraduate student researcher, Thomas Zemen joined Mastercard as a Software Engineer. Congrats, Thomas!
Dr. Imteaj joined as a Guest Editor of Privacy and Security in Federated Learning, A special issue of Journal of Surveillance, Security and Safety.
Our latest research on Label Flipping Data Poisoning Attack Against Wearable Human Activity Recognition System is accepted at IEEE Symposium Series on Computational Intelligence!