Xiaohui of University of Massachusetts, Boston

The increasing use of Online Large Language Models (OLLMs) for processing images has introduced significant privacy risks, as individuals frequently upload images for various utilities, unaware of the potential for privacy violations. Images contain relationships to Personally Identifiable Information (PII), where even seemingly harmless/unobjectionable details can indirectly reveal sensitive information through surrounding clues. This paper explores the critical issue of PII disclosure in images uploaded to OLLMs and its implications for user privacy. We investigate how the extraction of contextual relationships from images can lead to direct(explicit) or indirect(implicit) exposure of PII, significantly compromising personal privacy. Furthermore, we propose methods to safeguard privacy while preserving the intended utility of the images in LLM-based applications. Our evaluation demonstrates the efficacy of these techniques, highlighting the delicate balance between maintaining utility and protecting privacy in online image processing environments.

Biography:

Xiaohui Liang is an Associate Professor and Associate Chair with the Department of Computer Science at University of Massachusetts, Boston (UMB) where he leads the Mobile Computing and Privacy (MobCP) Lab. His research interests are Mobile Healthcare, Internet of Things, Wearable Computing, and Security and Privacy for Communication and Networking Systems. He has published over 100 refereed journal and conference papers. He received the Early Career Research Excellence Award from UMass Boston College of Science and Mathematics in 2020, the Early Career Award for Excellence in Scalable Computing from IEEE Technical Committee on Scalable Computing in 2017, the Best Land Transportation Paper Award from IEEE Vehicular Technology Society in 2017, the Internet of Things Technology Research Award from Google in 2016, the Best Paper Award in BodyNets 2010. His Google Scholar H-Index is 55 and total citation is 13,832 as of December 2025. Prior to joining UMB in Fall 2015, he was a Postdoctoral Researcher (with Professor David Kotz, FIEEE) at Dartmouth College in 2014-2015, and worked on an NSF frontier project THaW. He received the PhD degree (awarded Outstanding Achievement in Graduate Studies) in Electrical and Computer Engineering (under the supervision of Professor Sherman Shen, FIEEE and Professor Xiaodong Lin, FIEEE), University of Waterloo, Canada, in 2013.

Email:

Address:United States

Rongxing of Queen's University

Crowdsourced traffic monitoring provides real-time insights for urban planning and congestion management, but directly reporting users’ GPS coordinates poses serious privacy risks. We propose HSR-PRR, an accuracy-enhanced frequency estimation scheme under Local Differential Privacy (LDP) for traffic monitoring. By combining grid-based discretization with hash-based binning, the server partitions the domain and filters irrelevant reports, reducing variance and improving estimation accuracy. The scheme is highly communication-efficient, requiring only a 1-bit response per reporting user. We prove that reporting users satisfy ε-LDP, while non-reporting users incur minimal leakage, and the query condition achieves k-anonymity. Analysis and experiments show that HSR-PRR achieves higher accuracy and lower total communication overhead than PRR, making it a practical and scalable solution for large-scale, privacy-preserving traffic monitoring.

Biography:

Rongxing Lu is a full professor at Queen's School of Computing. Previously, he served as the Acting Director of the Canadian Institute for Cybersecurity (CIC), held the Mastercard IoT Research Chair, and was a professor in the Faculty of Computer Science at the University of New Brunswick (UNB). Before that, he was a tenure-track assistant professor at the School of Electrical and Electronic Engineering, Nanyang Technological University (NTU), Singapore, from April 2013 to August 2016. Rongxing Lu worked as a Postdoctoral Fellow at the University of Waterloo from May 2012 to April 2013. He was awarded the most prestigious “Governor General’s Gold Medal”, when he received his PhD degree from the Department of Electrical & Computer Engineering, University of Waterloo, Canada, in 2012; and won the 8th IEEE Communications Society (ComSoc) Asia Pacific (AP) Outstanding Young Researcher Award, in 2013. Dr. Lu is an IEEE Fellow. His research interests include IoT-Big Data security and privacy, privacy enhancing technologies, and applied cryptography. He has published extensively in his areas of expertise with H-index 97 and citations 43,700+ from Google Scholar as of January 2026, and was the recipient of 10 best (student) paper awards from some reputable journals and conferences. Dr. Lu served/ serves as the Chair of 2022-2023 IEEE ComSoc CIS-TC (Communications and Information Security Technical Committee), and the founding Co-chair of IEEE TEMS Blockchain and Distributed Ledgers Technologies Technical Committee (BDLT-TC). Dr. Lu is an IEEE ComSoc Distinguished Lecturer (2024-2025). Dr. Lu is the Winner of 2016-17 and 2023-24 Excellence in Teaching Award in FCS, UNB.

Email:

Address:Canada

Nan of University of Ottawa

Enhancing the resilience of fundamental infrastructures is crucial considering the increasingly frequent occurrence of natural disasters. Unmanned aerial vehicles (UAVs) have been extensively discussed as flexible and effective disaster rescue devices for the communication system, but their service quantity and quality are severely constrained by their limited battery capacities.

In this talk, we explore the potential of leveraging both the power provision and computational capabilities of electric vehicles (EVs) to provide prolonged rescue services for disaster-damaged areas across multiple dimensions. Specifically, a cooperative UAV-EV rescue framework is developed to characterize the cooperation procedure of UAV-EV pairs. Then, a two-tier matching problem is formulated where the lower tier maximizes the UAV operation period leveraging EVs’ computing and recharging services while the upper tier matches UAVs with EVs considering the optimized UAV operation period and statuses of different outage regions. We will present preliminary simulation results to validate the effectiveness of the proposed framework and discuss future work.

Biography:

Dr. Nan Chen is an Assistant Professor at the University of Ottawa, where she leads research on the planning and operation of electric vehicle charging infrastructure. Her research interests include exploring the potential of vehicle-to-grid technology to mitigate grid overloads and enhancing the resilience of fundamental infrastructures.

Email:

Address:Canada

Kuan of University of Nebraska-Lincoln

E-health allows smart devices and medical institutions to collaboratively collect patients' data, which is trained by Artificial Intelligence (AI) technologies to help doctors make diagnosis. By allowing multiple devices to train models collaboratively, federated learning is a promising solution to address the communication and privacy issues in e-health. However, directly applying federated learning in e-health faces many challenges. To this end, we provide a thorough study on an effective integration of HFL and VFL, to achieve communication efficiency and overcome the recent challenges when data is both horizontally and vertically partitioned. We present a hybrid federated learning framework with a Hybrid Stochastic Gradient Descent algorithm to train models. Then, we theoretically analyze the convergence upper bound of the proposed algorithm. Using the convergence results, we design adaptive strategies to adjust the training parameters and shrink the size of transmitted data.

Biography:

Kuan Zhang is an Associate Professor with the Department of Electrical and Computer Engineering, University of Nebraska–Lincoln, USA, where he was an Assistant Professor from 2017 to 2023. He received the Ph.D. degree in electrical and computer engineering from the University of Waterloo, Waterloo, ON, Canada, in 2016. Dr. Zhang has published over 100 papers in journals and conferences. His research interests include cyber security, AI, and cloud/edge computing. Dr. Zhang was a recipient of the Best Paper Award at IEEE WCNC 2013, Securecomm 2016, and ICC 2020. Dr. Zhang served as editor-in-chief of Encyclopedia of Wireless Networks. He is associate editor of IEEE Transactions on Wireless Communications, IEEE Internet of Things, IEEE Communications Survey and Tutorials, and Peer to Peer Networking and Application. He also served as symposium co-chair of IEEE ICC, CIC/ICCC.

Email:

Address:United States

Jianbing of Queen's University

Text-guided image manipulation with diffusion models enables flexible and precise editing based on prompts, but raises ethical and copyright concerns due to potential unauthorized modifications. To address this, we propose SecureT2I, a secure framework designed to prevent unauthorized editing in diffusion-based generative models. SecureT2I is compatible with both general-purpose and domain-specific models and can be integrated via lightweight fine-tuning without architectural changes. We categorize images into a permit set and a forbid set based on editing permissions. For the permit set, the model learns to perform high-quality manipulations as usual. For the forbid set, we introduce training objectives that encourage vague or semantically ambiguous outputs (e.g., blurred images), thereby suppressing meaningful edits. The core challenge is to block unauthorized editing while preserving editing quality for permitted inputs. To this end, we design separate loss functions that guide selective editing behavior. Extensive experiments across multiple datasets and models show that SecureT2I effectively degrades manipulation quality on forbidden images while maintaining performance on permitted ones. We also evaluate generalization to unseen inputs and find that SecureT2I consistently outperforms baselines. Additionally, we analyze different vagueness strategies and find that resize-based degradation offers the best trade-off for secure manipulation control.

Biography:

Jianbing Ni is currently an associate professor in the Department of Electrical and Computer Engineering at Queen's University, Canada, and the Tier-2 Canada Research Chair in Intelligent System Security and Privacy. His research interests include mobile network security, trustworthy artificial intelligence, cloud/edge computing security, and blockchain technology. He has published over 120 papers in IEEE Transactions and highly selected conferences. He also received the Best Paper Award for IEEE TMC in 2022, the IEEE Vehicular Technology Society Early Career Award in 2022, the IEEE Technical Committee on Scalable Computing Award for Excellence in Scalable Computing (Early Career Researchers) in 2023, the IEEE ComSoc CISTC 2024 Early Career Award, and IEEE Computer Society's 2025 Computing’s Top 30 Early Career Professionals. He is serving as an associate editor for IEEE T-IFS, IEEE T-DSC, and IEEE Systems Journal.

Email:

Address:Canada