The Second Workshop on Edge Intelligence in conjunction with ACM SEC 2025, Washington, D.C., USA, [December 3--6, 2025]
Edge intelligence is an emerging paradigm that brings AI model training and deployment closer to data sources—whether on devices, embedded systems, or edge computing nodes. By processing data locally, it reduces latency, energy consumption, and bandwidth demands while enhancing privacy and security through minimized data transmission. However, model composition and resource constraints are key challenges with respect to edge intelligence. Yet, edge intelligence is rapidly gaining attention for its transformative potential across IoT, autonomous systems, healthcare, smart cities, and precision agriculture. By enabling real-time analytics, autonomous decision-making, and efficient edge-cloud collaboration, it offers a compelling alternative to centralized cloud-dependent approaches.
This workshop invites contributions that explore the latest advances, challenges, and applications in edge intelligence, pushing the boundaries of what is possible at the network edge.
Topics of interests
- Cloud-edge collaborative training and inference
- Foundation models at the edge
- Gen AI for edge computing and edge computing for Gen AI
- LLM-driven edge network/compute policy management/optimization
- Federated learning via edge systems
- Edge-assisted artificial intelligence
- Efficient (deep) learning on edge devices
- Efficient (distributed) inference on end devices
- Lightweight container orchestration for domain-specific AI
- Applications of edge AI in extended reality (XR), cyber-physical systems (CPS), automation, healthcare, etc.
Only electronic submissions in PDF will be accepted. Submitted papers must be written in English and must be rendered without error using standard PDF viewing tools. Submitted papers must be no longer than 6 single-spaced 8.5'' x 11''; pages, including figures and tables, but excluding references, and using 10-point type on 12-point (single-spaced) leading, two-column format, Times Roman, or a similar font, within a text block 7.14'' wide x 9.22'' deep. IEEE Standard template for Latex and Word meet these specifications and can be found at: https://www.ieee.org/conferences/publishing/templates.html. Papers not meeting these criteria will be rejected without review, and no deadline extensions will be granted for reformatting. Pages should be numbered, and figures and tables should be legible in black and white, without requiring magnification.
Submission website: https://sec25-ei.hotcrp.com
At least one of the authors of each paper accepted must register for the workshop.
Please note that camera-ready papers must conform to the most recent ACM template, including the bibliographic and copyright strip. For more details, please refer to the camera-ready instructions page.
Important Dates
Submission Deadline: September 15, 2025 October 4, 2025 (Firm)
Notification: October 24, 2025
Camera-ready paper deadline: November 1, 2025
Workshop date: December 6, 2025
Organization
Edge Intelligence Workshop 2025 will be a half-day workshop with two paper presentation sessions, one keynote, and potentially a poster / short-paper presentation session.
Workshop Organizers
TPC co-chairs
Prasad Calyam, University of Missouri-Columbia, USA
Xiaoli Liu, University of Helsinki, Finland
Jaya Prakash Champati, University of Victoria, Canada
Organizing Chair
James Gross, KTH Stockholm
Technical Program Committee
Dianlei Xu, University of Helsinki
Hei Victor Cheng, Aarhus University
James Gross, KTH Royal Institute of Technology
Jaya Prakash Champati, University of Victoria
Jianli Pan, George Mason University
Roberto Morabito, EURECOM
Nikhil Pratap Ghanathe, University of British Columbia
Paul Pop, Technical University of Denmark
Prasad Calyam, University of Missouri-Columbia
Qi Zhang, Aarhus University
Saptarshi Debroy, City University of New York
Xiang Su, University of Helsinki
Xiaoli Liu, University of Helsinki
Program
10:30 - 10:50
How Heavy is the Edge? Resource Utilization of Edge Generative AI on Distributed AI Infrastructure
Liang Wu (University of Wisconsin-Milwaukee), Zhen Zeng (University of Wisconsin-Milwaukee), Zhongshu Gu (IBM Research), and Pengxia Wu (Rockwell Automation)
10:50 - 11:10
Performance Evaluation of Whisper-Series Speech Transcription Models on Raspberry Pi
Yue Cao (Syracuse University)
11:10 - 11:30
Edge-Enabled Scalable Routing via Graph Neural Network Pruning and Metaheuristic Optimization
Subrahmanya Chandra Bhamidipati (University of Missouri-Columbia), William Echols (University of California Berkeley), Jesus Lopez Olivares (San Joaquin Delta College), Kenzie Markley (Arizona State University), Sharan Srinivas (University of Missouri-Columbia), and Prasad Calyam (University of Missouri-Columbia)
11:30 - 11:50
RepVal: A Skeleton-based Validation System for Functional Fitness Repetition on Edge Devices
Lucas Alves (Oakland University), Fan Li (Oakland University), and Lanyu Xu (Oakland University)
11:50
Closing remarks