paperIEEE Xplore·1 Nov 2026·Zhi Yao, Zhiqing Tang, Hanshuai Cui, Weijia Jia, Wei Zhao
DIAMM: A Distributed Inference Acceleration Framework for Multimodal Edge Applications
DIAMM splits multimodal model inference across edge devices to cut latency.
Extract
IEEE Transactions on Parallel and Distributed Systems