发表机构
Advanced Research Lab, NavInfo Europe; Department of Mathematics and Computer Science, Eindhoven University of Technology（先进研究实验室，欧洲四维图新; 埃因霍温理工大学数学与计算机科学系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究受人类视觉系统启发，提出中央凹引导动态变换器（FDT），通过注视和中央凹模块实现动态令牌选择。FDT对噪声和攻击有强恢复力，在50%注视预算下，比DeiT-S准确率高且乘加运算减少，是迈向结合自适应计算与恢复能力的人工神经网络的重要一步。
AI 中文摘要
人类视觉系统采用中央凹采样和眼球运动来实现高效感知，节省代谢能量和计算资源。受此稳健性和适应性启发，我们引入了中央凹引导动态变换器（FDT），这是一种将这些机制集成到视觉变换器框架中的中央凹引导动态令牌选择架构。FDT对各种类型的噪声和对抗攻击具有很强的恢复能力，尽管没有针对此类挑战进行明确训练。这种内在的稳健性通过使用注视和中央凹模块来实现：注视模块识别注视点以过滤掉无关信息，而中央凹模块生成具有多尺度信息的中央凹嵌入。在50%注视预算设置下，FDT比DeiT-S实现了更高的准确率（81.9%对80.9%），同时将乘加运算减少了34.57%，突出了其在准确率-效率权衡上的一个工作点。这些特性使FDT成为迈向将自适应计算与提高恢复能力相结合的人工神经网络的受人类视觉系统启发的一步。
英文摘要
The human visual system (HVS) employs foveated sampling and eye movements to achieve efficient perception, conserving both metabolic energy and computational resources. Drawing inspiration from this robustness and adaptability, we introduce the Foveated Dynamic Transformer (FDT), a foveation-guided dynamic token-selection architecture that integrates these mechanisms into a vision transformer framework. The FDT exhibits strong resilience to various types of noise and adversarial attacks, despite not being explicitly trained for such challenges. This inherent robustness is achieved through the use of fixation and foveation modules: the fixation module identifies fixation points to filter out irrelevant information, while the foveation module generates foveated embeddings with multi-scale information. At the 50% fixation-budget setting, FDT achieves higher accuracy than DeiT-S (81.9% vs. 80.9%) while reducing multiply-accumulate operations by 34.57%, highlighting one operating point on its accuracy-efficiency trade-off. These attributes position FDT as an HVS-inspired step toward artificial neural networks that combine adaptive computation with improved resilience.
