DeepScan: 一种无需训练的框架，用于大视觉-语言模型中的视觉引导推理
DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models
AI总结：
DeepScan是一种无需训练的框架，通过层次扫描、聚焦和证据增强推理提升大视觉-语言模型在视觉任务中的表现。
摘要：
人类能够在嘈杂环境中通过识别关键线索并以自底向上的方式将它们与完整上下文相关联，从而稳健地定位视觉证据并提供基于现实的答案。受此启发，我们提出了DeepScan，一种无需训练的框架，该框架结合了层次扫描、聚焦和证据增强推理，用于大视觉-语言模型（LVLMs）中的视觉引导推理。与现有方法追求完整证据的一次性定位不同，层次扫描通过局部线索探索和多尺度证据提取，以自底向上的方式恢复证据，有效减轻了干扰上下文的影响。聚焦然后通过LVLMs和视觉专家的协作优化局部化证据视图。最后，证据增强推理通过混合证据记忆聚合多粒度视图，从而产生准确且可解释的答案。实验结果表明，DeepScan在多种视觉任务中显著提升了LVLMs，特别是在细粒度视觉理解方面。当与Qwen2.5-VL-7B结合时，其在V*上的整体准确率达到90.6%。此外，DeepScan在各种架构和模型规模上为LVLMs提供了一致的改进，而无需额外的适应成本。
英文摘要：
Humans can robustly localize visual evidence and provide grounded answers even in noisy environments by identifying critical cues and then relating them to the full context in a bottom-up manner. Inspired by this, we propose DeepScan, a training-free framework that combines Hierarchical Scanning, Refocusing, and Evidence-Enhanced Reasoning for visually grounded reasoning in Large Vision-Language Models (LVLMs). Unlike existing methods that pursue one-shot localization of complete evidence, Hierarchical Scanning performs local cue exploration and multi-scale evidence extraction to recover evidence in a bottom-up manner, effectively mitigating the impacts of distractive context. Refocusing then optimizes the localized evidence view through collaboration of LVLMs and visual experts. Finally, Evidence-Enhanced Reasoning aggregates multi-granular views via a hybrid evidence memory and yields accurate and interpretable answers. Experimental results demonstrate that DeepScan significantly boosts LVLMs in diverse visual tasks, especially in fine-grained visual understanding. It achieves 90.6% overall accuracy on V* when integrated with Qwen2.5-VL-7B. Moreover, DeepScan provides consistent improvements for LVLMs across various architectures and model scales without additional adaptation cost.
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。
