头感知视觉裁剪：通过注意力引导的子图像增强细粒度VQA
Head-Aware Visual Cropping: Enhancing Fine-Grained VQA with Attention-Guided Subimage
AI总结：
通过注意力引导的子图像增强细粒度VQA，提出无需训练的HAVC方法，利用优化的注意力头提升视觉定位精度。
摘要：
多模态大语言模型（MLLMs）在视觉问答（VQA）中表现出强大的性能，但由于低分辨率输入和噪声注意力聚合，在细粒度推理方面仍存在局限。我们提出了Head Aware Visual Cropping（HAVC），一种无需训练的方法，通过利用选择性优化的注意力头子集来提高视觉定位。HAVC首先通过基于OCR的诊断任务过滤头，确保仅保留那些具有真实定位能力的头。在推理过程中，这些头进一步通过空间熵进行强化，以获得更强的空间集中性，并通过梯度敏感性进行预测贡献优化。融合的信号产生一个可靠的视觉裁剪指导图，该图突出显示最相关的任务区域，并指导随后提供的子图像与图像-问题对一起进行裁剪。在多个细粒度VQA基准上的广泛实验表明，HAVC在多个细粒度VQA基准上一致优于最先进的裁剪策略，实现了更精确的定位，更强的视觉定位，提供了一种简单而有效的策略，以提高MLLMs的精度。
英文摘要：
Multimodal Large Language Models (MLLMs) show strong performance in Visual Question Answering (VQA) but remain limited in fine-grained reasoning due to low-resolution inputs and noisy attention aggregation. We propose \textbf{Head Aware Visual Cropping (HAVC)}, a training-free method that improves visual grounding by leveraging a selectively refined subset of attention heads. HAVC first filters heads through an OCR-based diagnostic task, ensuring that only those with genuine grounding ability are retained. At inference, these heads are further refined using spatial entropy for stronger spatial concentration and gradient sensitivity for predictive contribution. The fused signals produce a reliable Visual Cropping Guidance Map, which highlights the most task-relevant region and guides the cropping of a subimage subsequently provided to the MLLM together with the image-question pair. Extensive experiments on multiple fine-grained VQA benchmarks demonstrate that HAVC consistently outperforms state-of-the-art cropping strategies, achieving more precise localization, stronger visual grounding, providing a simple yet effective strategy for enhancing precision in MLLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
