Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2604.15622 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917456184868864 |
|---|---|
| author | Zhao, Yiwei Zheng, Yi Su, Huapeng Lin, Jieyu Ambrogio, Stefano Jose, Cijo Ramamonjisoa, Michael Labatut, Patrick De Salvo, Barbara Liu, Chiao Gibbons, Phillip B. Li, Ziyun |
| author_facet | Zhao, Yiwei Zheng, Yi Su, Huapeng Lin, Jieyu Ambrogio, Stefano Jose, Cijo Ramamonjisoa, Michael Labatut, Patrick De Salvo, Barbara Liu, Chiao Gibbons, Phillip B. Li, Ziyun |
| contents | Language-aligned vision foundation models (VFMs) enable versatile visual understanding for always-on contextual AI, but their deployment on edge devices is hindered by strict latency and power constraints. We present AdaVFM, an adaptive framework for efficient on-device inference of language-aligned VFMs that dynamically adjusts computation based on scene context and task complexity. Our key insight is that the effect of model size reduction on performance is task-dependent in vision applications, motivating a runtime-adaptive execution strategy. AdaVFM integrates neural architecture search (NAS) into the language-aligned VFM backbone to enable lightweight subnet execution during runtime. A multimodal large language model (LLM) deployed on the cloud enables runtime control with a context-aware agent. This synergy allows efficient model adaptation under diverse conditions while maintaining strong accuracy. Extensive experiments on zero-shot classification and open-vocabulary segmentation demonstrate that AdaVFM achieves state-of-the-art accuracy-efficiency trade-offs, surpassing prior baselines by up to $7.9\%$ in acc@1 on IN1K and $5.2\%$ mIoU on ADE20K over the best models of comparable VFM sizes. For models with similar accuracy, AdaVFM further reduces average FLOPs by up to $77.9\%$. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_15622 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AdaVFM: Adaptive Vision Foundation Models for Edge Intelligence via LLM-Guided Execution Zhao, Yiwei Zheng, Yi Su, Huapeng Lin, Jieyu Ambrogio, Stefano Jose, Cijo Ramamonjisoa, Michael Labatut, Patrick De Salvo, Barbara Liu, Chiao Gibbons, Phillip B. Li, Ziyun Computer Vision and Pattern Recognition Machine Learning Language-aligned vision foundation models (VFMs) enable versatile visual understanding for always-on contextual AI, but their deployment on edge devices is hindered by strict latency and power constraints. We present AdaVFM, an adaptive framework for efficient on-device inference of language-aligned VFMs that dynamically adjusts computation based on scene context and task complexity. Our key insight is that the effect of model size reduction on performance is task-dependent in vision applications, motivating a runtime-adaptive execution strategy. AdaVFM integrates neural architecture search (NAS) into the language-aligned VFM backbone to enable lightweight subnet execution during runtime. A multimodal large language model (LLM) deployed on the cloud enables runtime control with a context-aware agent. This synergy allows efficient model adaptation under diverse conditions while maintaining strong accuracy. Extensive experiments on zero-shot classification and open-vocabulary segmentation demonstrate that AdaVFM achieves state-of-the-art accuracy-efficiency trade-offs, surpassing prior baselines by up to $7.9\%$ in acc@1 on IN1K and $5.2\%$ mIoU on ADE20K over the best models of comparable VFM sizes. For models with similar accuracy, AdaVFM further reduces average FLOPs by up to $77.9\%$. |
| title | AdaVFM: Adaptive Vision Foundation Models for Edge Intelligence via LLM-Guided Execution |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2604.15622 |