发表机构
Qualcomm AI Research（高通人工智能研究中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究从视觉状态跟踪和长度泛化角度，探讨局部、顺序视觉模型是否有计算优势。受语言模型启发，研究简单视觉任务中视觉模型行为。发现其会利用全局捷径，基于严格局部感知的策略可缓解此问题，表明局部注意力对稳健组合泛化很关键。
AI 中文摘要
人类视觉系统的一个显著特征是通过一系列局部中央凹瞥见摄取视觉信息，而非单一全局计算，这与当今多数流行的计算机视觉模型不同。本文从视觉状态跟踪和长度泛化角度研究局部、顺序视觉模型是否有计算优势。受语言模型中长度泛化研究启发，对在需要聚合图像局部信息的简单视觉任务上训练的视觉模型行为进行研究。实验表明，视觉模型会利用全局捷径而无法在任务长度或复杂度上泛化，基于严格局部感知的循环视觉策略可缓解此问题，结果显示局部注意力可能是稳健组合泛化的关键且被忽视的要求。
英文摘要
A striking feature of the human visual system is that it ingests visual information through a series of local foveated glimpses, rather than a single global computation. This makes human vision distinctly different from most popular computer vision models in use today, which input images globally and in a single shot. A natural question therefore is whether local, sequential vision models may provide any fundamental computational benefits in addition to being biologically more plausible than global models. In this work, we investigate this question from the perspective of visual state tracking and length generalization. Inspired by recent studies of length generalization in language models, we study the behavior of vision models trained on simple vision tasks that require the aggregation of local information across an image. Our experiments reveal that, similar to language models, vision models can learn to exploit global shortcuts and thereby fail to generalize over task length or complexity. We also show that recurrent vision policies based on strictly local perception can mitigate these failures, thereby allowing models to generalize on these tasks. Our results show that local attention may be an essential overlooked requirement for robust compositional generalization.
CommentsAccepted at ECCV 2026
