一项新研究表明，边缘设备上视觉语言模型（VLM）的能源消耗主要由生成的输出量驱动，而非视觉输入的复杂性。研究人员发现，生成每个输出令牌比处理输入令牌花费的时间更长，消耗的能源也更多。这表明，减少VLM能源消耗的努力应侧重于控制输出长度，因为即使消除视觉输入处理，能源节省也微乎其微。 AI
影响 专注于VLM的输出长度控制可以显著降低边缘设备的能源消耗。
排序理由 该集群包含一篇详细介绍VLM能源消耗研究结果的论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2607.09520v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood. Existing efficiency efforts focus predominantly on reducing visual tokens, implicitly…
Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood. Existing efficiency efforts focus predominantly on reducing visual tokens, implicitly treating visual processing as the dominant energy…
Convolutional neural networks (CNNs) have demonstrated encouraging results in image classification tasks. However, the prohibitive computational cost of CNNs hinders the deployment of CNNs onto resource-constrained embedded devices. To address this issue, we propose EdgeCompress,…