AdaVFM integrates neural architecture search into vision foundation model backbones and uses a cloud multimodal LLM agent to enable runtime-adaptive lightweight subnet execution, delivering up to 7.9% higher accuracy and 77.9% lower FLOPs than fixed-size baselines on edge devices.
In: 2016 IEEE 9th Workshop on Software En- gineering and Architectures for Realtime Interactive Systems (SEARIS)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
AdaVFM: Adaptive Vision Foundation Models for Edge Intelligence via LLM-Guided Execution
AdaVFM integrates neural architecture search into vision foundation model backbones and uses a cloud multimodal LLM agent to enable runtime-adaptive lightweight subnet execution, delivering up to 7.9% higher accuracy and 77.9% lower FLOPs than fixed-size baselines on edge devices.