PIX-TAB：结合投机解码与区域图像分割的高效像素级精确表格结构识别方法
PIX-TAB: Efficient PIXel-Precise TABle Structure Recognition Approach with Speculative Decoding and Region-Based Image Segmentation
位置感知的像素级 token 与仅训练期使用的框监督，为文档版面这类密集结构化预测提供了可复用的空间定位方案；投机解码在序列式结构生成上的加速也可迁移到其他自回归式视觉输出任务。
原文依据 position-aware pixel-precise tokens for deterministic cell reconstruction; speculative decoding for faster sequence generation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论文档 AI 中的表格结构识别面临布局不一致、合并单元格与复杂嵌套结构等显著挑战，而大规模、多样化标注数据的稀缺进一步加剧了这一问题。本文提出 PIX-TAB（高效像素级精确表格结构识别方法），借助可在设备端运行的小型轻量模型提供精确的像素级结构。该方法与语言无关，只需替换光学字符识别（OCR）模型即可新增对一种语言的支持，而无需修改核心结构识别模型。关键创新包括：用于确定性单元格重建的位置感知像素级精确 token；用于加速序列生成的投机解码；用于稳定空间定位的仅训练期框监督；基于区域的图像分割。为缓解数据稀缺，我们提出了一套大规模合成表格数据集的生成流程。实验结果验证了各组成部分的有效性。为弥补现有评测方法的不足，我们引入 TEDS_struct100 与 TEDS_100 两项指标。投机解码方法在保持精度的同时显著提升了识别速度。最终，上述技术组合得到一个面向移动端优化的模型，其速度比完整规模版本快三倍以上。
英文原文摘要 1368 字符
Table structure recognition in document AI faces significant challenges due to layout inconsistencies, merged cells, and complex nested structures, which is further exacerbated by the scarcity of large, diverse annotated datasets. In this paper, we present PIX-TAB (Efficient PIXel-Precise TABle Structure Recognition Approach) that provides exact, pixel-level structure using a small, lightweight model that can run on-device. The approach is language-agnostic, as it allows adding support for a new languages simply by replacing the Optical Character Recognition (OCR) model without modifying to the core structure recognition model. Key innovations include: position-aware pixel-precise tokens for deterministic cell reconstruction; speculative decoding for faster sequence generation, and training-only box supervision to stabilize spatial grounding; region-based image segmentation. To mitigate data scarcity we propose a pipeline for generating a large synthetic table dataset. Experimental results validate each component. To address the limitations of existing evaluation methods we introduce TEDS_ struct100 and TEDS_ 100 metrics. Speculative decoding approach significantly improves recognition speed while maintaining accuracy. Finally, the combined techniques enable a mobile-optimized model that is more than three times faster than the full-size version.