研究人员开发了一种新的视觉文档理解方法，该方法无需基于坐标的区域标签。通过比较坐标接口和基于文本的引用检索管道，他们发现后者显著提高了证据召回率，并降低了多个视觉语言模型中的幻觉率。这种方法使用 GRPO 配方来训练模型，使其能够引用更好的证据，而无需昂贵的区域级监督，并在 8B 主干模型上展示了严格归因准确性的提高。 AI
影响 这项研究提供了一种更有效、更准确的视觉文档理解方法，有望改进 AI 模型处理和引用文档信息的方式。
排序理由 详细介绍视觉文档理解新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2607.24651v1 Announce Type: cross Abstract: Reliable visual document understanding requires a model to attribute each answer to the evidence regions that support it. Recent benchmarks and systems express this step through a coordinate interface: the model outputs the coordi…
Reliable visual document understanding requires a model to attribute each answer to the evidence regions that support it. Recent benchmarks and systems express this step through a coordinate interface: the model outputs the coordinates of bounding boxes that mark the evidence reg…
Reliable visual document understanding requires a model to attribute each answer to the evidence regions that support it. Recent benchmarks and systems express this step through a coordinate interface: the model outputs the coordinates of bounding boxes that mark the evidence reg…