提出首个复杂公式识别数据集与细节聚焦网络，显著提升公式解析准确率。
Enhancing Complex Formula Recognition with Hierarchical Detail-Focused Network
- 构建分层细节聚焦网络，逐级解析公式结构与细微差异。
- 在百万级数据上训练，测试集包含多解释复杂公式，验证模型鲁棒性。
- 适用于数学文档数字化、教育AI等需要高精度公式理解的场景。
由于公式可能存在多种解释，层次化且复杂的数学表达式识别（MER）极具挑战性，导致解析与评估困难。本文提出首个专为此问题设计的数据集——分层细节聚焦识别数据集（HDR），包含大规模训练集HDR-100M（一亿样本）和用于全面评估模型性能的测试集HDR-Test，后者包含多个复杂分层公式的不同解释。此外，公式解析常因细粒度错误而受影响。为此，我们提出分层细节聚焦识别网络（HDNet），其引入分层子公式模块，专注精确处理公式细节，显著提升MER性能。实验表明，HDNet在多个数据集上均优于现有模型。
原文摘要 · Abstract (English)
Hierarchical and complex Mathematical Expression Recognition (MER) is challenging due to multiple possible interpretations of a formula, complicating both parsing and evaluation. In this paper, we introduce the Hierarchical Detail-Focused Recognition dataset (HDR), the first dataset specifically designed to address these issues. It consists of a large-scale training set, HDR-100M, offering an unprecedented scale and diversity with one hundred million training instances. And the test set, HDR-Test, includes multiple interpretations of complex hierarchical formulas for comprehensive model performance evaluation. Additionally, the parsing of complex formulas often suffers from errors in fine-grained details. To address this, we propose the Hierarchical Detail-Focused Recognition Network (HDNet), an innovative framework that incorporates a hierarchical sub-formula module, focusing on the precise handling of formula details, thereby significantly enhancing MER performance. Experimental results demonstrate that HDNet outperforms existing MER models across various datasets.
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。