TDATR：通过表格细节感知学习和单元格级视觉对齐改进端到端表格识别
TDATR: Improving End-to-End Table Recognition via Table Detail-Aware Learning and Cell-Level Visual Alignment
论文提出的结构引导的单元格定位模块和单元格级视觉-语言对齐方法，可被其他需要细粒度视觉-语言对齐的文档理解任务（如表格问答、文档布局分析）借鉴；其“先感知后融合”策略和利用多任务学习增强数据效率的思路，对数据受限场景下的视觉-语言模型训练有参考价值。
原文依据 we design a structure-guided cell localization module integrated into the end-to-end TR framework, which efficiently locates cell and strengthens vision-language alignment.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论表格广泛存在于各类文档中，使得表格识别（TR）成为文档分析中的一项基础任务。现有的模块化表格识别流程分别对表格结构和内容进行建模，导致集成效果欠佳且工作流程复杂。端到端方法严重依赖大规模表格识别数据，在数据受限场景下表现不佳。为解决这些问题，我们提出TDATR（表格细节感知的表格识别），通过表格细节感知学习和单元格级视觉对齐来改进端到端表格识别。TDATR采用“先感知后融合”的策略。模型首先执行表格细节感知学习，通过语言建模范式下设计的多个结构理解和内容识别任务，联合感知表格结构和内容。这些任务能够自然利用来自不同场景的文档数据，以增强模型的鲁棒性。随后，模型整合隐式表格细节以生成结构化HTML输出，从而在有限数据训练时实现更高效的表格识别建模。此外，我们设计了一个结构引导的单元格定位模块，集成到端到端表格识别框架中，该模块能高效定位单元格并加强视觉-语言对齐，提升了表格识别的可解释性和准确性。我们在七个基准上无需针对数据集微调即达到最优或极具竞争力的性能。
英文原文摘要 1392 字符
Tables are pervasive in diverse documents, making table recognition (TR) a fundamental task in document analysis. Existing modular TR pipelines separately model table structure and content, leading to suboptimal integration and complex workflows.End-to-end approaches rely heavily on large-scale TR data and struggle in data-constrained scenarios.To address these issues, we propose TDATR (Table Detail-Aware Table Recognition) improves end-to-end TR through table detail-aware learning and cell-level visual alignment.TDATR adopts a "perceive-then-fuse" strategy. The model first performs table detail-aware learning to jointly perceive table structure and content through multiple structure understanding and content recognition tasks designed under a language modeling paradigm. These tasks can naturally leverage document data from diverse scenarios to enhance model robustness.The model then integrates implicit table details to generate structured HTML outputs, enabling more efficient TR modeling when trained with limited data.Furthermore, we design a structure-guided cell localization module integrated into the end-to-end TR framework, which efficiently locates cell and strengthens vision-language alignment. It enhances the interpretability and accuracy of TR.We achieve state-of-the-art or highly competitive performance on seven benchmarks without dataset-specific fine-tuning.