TRivia: 基于视觉-语言模型的自监督微调用于表格识别
TRivia: Self-supervised Fine-tuning of Vision-Language Models for Table Recognition
AI总结：
TRivia通过自监督微调方法，使预训练视觉-语言模型直接从未标注的表格图像中学习表格识别，无需人工标注，提升了表格识别性能。
摘要：
表格识别（TR）旨在将表格图像转换为半结构化表示，如HTML或Markdown。作为文档解析的核心组件，TR长期以来依赖于监督学习，近期主要通过使用标注数据对视觉-语言模型（VLMs）进行微调。尽管VLMs将TR推至新高度，但进一步提升性能需要大量标注数据，这成本高昂。因此，尽管专有模型不断推动性能边界，开源模型往往受限于资源有限，且由于隐私法规，许多情况下仍是唯一可行选项。为此，我们引入TRivia，一种自监督微调方法，使预训练VLMs能直接从野外的未标注表格图像中学习TR。基于组相对策略优化，TRivia自动识别最有效促进学习的未标注样本，并通过基于问答的奖励机制消除人工标注需求。注意力引导模块为每个表格图像生成多样化问题，而能够解释识别结果并正确回答问题的能力提供反馈以优化TR模型。这种闭环过程使TR模型能够自主学习识别、结构化和推理性表格，而无需标注数据。利用此流程，我们提出TRivia-3B，一个开源、紧凑且最先进的TR模型，在三个流行基准上超越现有系统（如Gemini 2.5 Pro、MinerU2.5）。模型和代码已发布：https://github.com/HKU-TASR/TRivia
英文摘要：
Table recognition (TR) aims to transform table images into semi-structured representations such as HTML or Markdown. As a core component of document parsing, TR has long relied on supervised learning, with recent efforts dominated by fine-tuning vision-language models (VLMs) using labeled data. While VLMs have brought TR to the next level, pushing performance further demands large-scale labeled data that is costly to obtain. Consequently, although proprietary models have continuously pushed the performance boundary, open-source models, often trained with limited resources and, in practice, the only viable option for many due to privacy regulations, still lag far behind. To bridge this gap, we introduce TRivia, a self-supervised fine-tuning method that enables pretrained VLMs to learn TR directly from unlabeled table images in the wild. Built upon Group Relative Policy Optimization, TRivia automatically identifies unlabeled samples that most effectively facilitate learning and eliminates the need for human annotations through a question-answering-based reward mechanism. An attention-guided module generates diverse questions for each table image, and the ability to interpret the recognition results and answer them correctly provides feedback to optimize the TR model. This closed-loop process allows the TR model to autonomously learn to recognize, structure, and reason over tables without labeled data. Leveraging this pipeline, we present TRivia-3B, an open-sourced, compact, and state-of-the-art TR model that surpasses existing systems (e.g., Gemini 2.5 Pro, MinerU2.5) on three popular benchmarks. Model and code are released at: https://github.com/HKU-TASR/TRivia
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。
