构建真实化学结构识别基准，解决图像复杂下的识别难题
MolRecBench-Wild: A Real-World Benchmark for Optical Chemical Structure Recognition
- 提出双维度难度框架，细分为37类视觉与化学挑战
- 创建5029个真实论文图像的基准数据集，覆盖全难度范围
- 设计新语义表示语言，支持非标准化学符号表达
光学化学结构识别（OCSR）旨在将科学文献中的分子图转化为机器可读格式，但现有系统在真实图像上仍不可靠，因视觉与化学复杂性显著。我们提出MOSAIC，一种包含37个细粒度标签的双维度难度框架，联合刻画分子图中的视觉干扰与化学语义挑战。基于此框架，构建了包含5029个结构的MolRecBench-Wild基准，数据来自820篇近期化学论文，覆盖真实出版物中观察到的完整难度谱。为实现超越SMILES和MolFile的忠实语义评估，我们提出CARBON，一种能表达价态变化、图标基团等非标准化学语义的表示语言。同时采用双轨评估协议，支持CARBON与SMILES输出以保障模型兼容性。对18个具备OCSR能力的模型进行综合实验，结果显示其在MolRecBench-Wild上性能严重下降，揭示了以往专利基准与真实学术场景间的巨大差距。
原文摘要 · Abstract (English)
Optical Chemical Structure Recognition (OCSR) aims to translate molecular diagrams in scientific literature into machine-readable formats, but current systems remain unreliable on real-world images due to substantial visual and chemical complexity. We introduce MOSAIC, a dual-dimensional difficulty framework with 37 fine-grained labels that jointly characterize visual interference and chemical semantic challenges in molecular diagrams. Based on this framework, we construct MolRecBench-Wild, a benchmark of 5,029 structures from 820 recent chemistry papers, covering the full difficulty spectrum observed in real publications. To enable faithful semantic evaluation beyond SMILES and MolFile, we propose CARBON, a representation language capable of expressing valence variations, icon-based groups, and other non-standard chemical semantics. We further adopt a dual-track evaluation protocol supporting both CARBON and SMILES outputs for broad model compatibility. Comprehensive experiments over 18 OCSR-capable models reveal severe performance degradation on MolRecBench-Wild, exposing a large gap between previous patent benchmarks and real-world academic scenarios.
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。