Edge and Mobile Foundation Models (EdgeFM 2025)

Abstract: With the recent democratization of AI, there is a pressing need of supporting AI on mobile and embedded devices at the edge, to allow intelligent and prompt decision making autonomously on these devices. To meet the devices’ constraints in computing capacity, current software solutions to on-device AI reduce the ML model’s complexity, but have major weaknesses in adapting to the changes of online data patterns and environmental contexts, resulting in significant reduction of model performance in difficult learning tasks. In this talk, I will present our recent research on achieving such full runtime adaptability, as a key enabler for prevalence of on-device AI in practical systems. I will first present how we leverage explainability in AI to adaptively involve the most appropriate model structures for on-device computations, and then elaborate on how we utilize such explainability to support real-time inference, runtime training and LLM fine-tuning on devices with extreme resource constraints.

Biography: Wei Gao is currently an Associate Professor in the Department of Electrical and Computer Engineering, University of Pittsburgh. His research interests lie in the intersection between AI, computer systems and computer vision, with a focus on the design and deployment of on-device AI models and algorithms on mobile, embedded and networked devices. He also has strong interests in applying the computationally efficient AI models into practical application domains to make societal impacts and benefit the human welfare. He has published more than 100 research papers at top AI, system and vision conference venues, including ICLR, AAAI, CVPR, ICCV, ASPLOS, MobiCom, MobiSys, SenSys, etc, and received multiple best paper awards or nominations.

Liang Li, Wen Wu (Pengcheng Laboratory)

Huajun Bai (Tsinghua University, Advanced Micro Devices, Inc.), Zihao An (Advanced Micro Devices, Inc.), Qing Wang (Tsinghua University), Ziqiong Liu (Advanced Micro Devices, Inc.), Dong Li (Advanced Micro Devices, Inc.), Emad Barsoum (Advanced Micro Devices, Inc.), Jiwu Shu (Tsinghua University)

Qingzhe Guo, Tu Ouyang, An Wang (Case Western Reserve University) [Online]

Zixu Hao (Tsinghua University), Ju Ren (Tsinghua University), Ting Cao (Institute of AI Industry Research, Tsinghua University)

Guohong Liu (Institute for AI Industry Research (AIR), Tsinghua University), Jialei Ye (University of Electronic Science and Technology of China), Jiacheng Liu (School of Computer Science, Peking University), Yuanchun Li (Institute for AI Industry Research (AIR), Tsinghua University), Wei Liu (MiLM Plus, Xiaomi Inc.), Pengzhi Gao (MiLM Plus, Xiaomi Inc.), Jian Luan (MiLM Plus, Xiaomi Inc.), Yunxin Liu (Institute for AI Industry Research (AIR), Tsinghua University)

Junlin Liu (Peking University), Yifeng Cai (Peking University), Ziqi Zhang (UIUC), Yao Guo (Peking University), Ding Li (Peking University) [Online]

Runxi Huang, Xiaomin Ouyang (Hong Kong University of Science and Technology)

Li Zhang, Mengwei Xu (Beijing University of Posts and Telecommunications)

Bernard Yap (University of Michigan, Ann Arbor), Zheng Li (University of Michigan, Ann Arbor), Jiasi Chen (University of Michigan, Ann Arbor)

Simon Jonas Bührer (ETH Zurich), Andreas Lindhardt Plesner (ETH Zurich), Till Aczel (ETH Zurich), Roger Wattenhofer (ETH Zurich)

Hannah B. Pasandi (University of California, Berkeley), Zhaowei Tan (University of California, Riverside), Mohammad Hosseini (Shahid Beheshti University), Sina Darabi (USI Lugano), Franck Rousseau (Grenoble INP), Juan A. Fraire (Inria-Lyon)

Hannah B. Pasandi (University of California, Berkeley)