FastTab用小模块和1D Transformer快速准确识别表格结构。
FastTab: A Fast Table Recognizer with a Tiny Recursive Module and 1D Transformers
- 用轻量递归模块+轴向1D Transformer实现全局推理与长程依赖建模。
- 在4个数据集上达到竞品水平，推理延迟低，支持跨行跨列合并单元格。
- 适合需要实时处理的文档结构化场景，如扫描件、复杂表格解析。
表格结构识别（TSR）需兼顾表级一致性（行列数、标题、跨行跨列）与精确分隔线定位。本文提出FastTab，一种以网格为中心的TSR模型，通过结合（i）轻量级微型递归模块（TRM）进行全局推理，以及（ii）轴向1D Transformer编码器捕捉行列方向上的长程依赖，避免了自回归式HTML解码。模型先预测行列数、标题行与分隔线构建网格，再利用区域对齐的细胞特征推断跨行/跨列。在四个基准数据集（PubTabNet、FinTabNet、PubTables-1M、SciTSR）上表现优异，且推理延迟极低。进一步研究了像素级匿名化下的鲁棒性，并扩展至支持相机拍摄文档中的弯曲分隔线。代码将公开于https://github.com/hamdilaziz/FastTab。
原文摘要 · Abstract (English)
Table structure recognition (TSR) requires both table-level coherence (row/column counts, headers, spanning cells) and precise separator localization. We introduce FastTab, a grid-centric TSR model that avoids autoregressive HTML decoding by combining (i) a lightweight Tiny Recursive Module (TRM) for global reasoning and (ii) axial 1D Transformer encoders that capture long-range dependencies along rows and columns. The model predicts row/column counts, header rows, and separators to construct a grid, then infers rowspan/colspan using ROI-aligned cell features. Across four benchmarks (PubTabNet, FinTabNet, PubTables-1M, and SciTSR), FastTab achieves competitive structure recovery performance while operating at low-latency inference. We further study robustness under pixel-level anonymisation and show an extension to curved separators for camera-captured documents. The source code will be made publicly available at https://github.com/hamdilaziz/FastTab .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。