MinerU.Chem: A High-Precision System for Optical Chemical Structure and Reaction Recognition

2026-08-04

AI Summary

This study addresses the challenge of accurately extracting molecular structures and reaction schemes from organic chemistry literature, a task that general-purpose document parsing systems struggle with, thereby hindering AI-driven chemical research. Building upon the MinerU framework, the authors propose CARBON—a novel representation scheme—and integrate five specialized modules for chemical relevance filtering, molecular detection, identifier extraction, and reaction diagram parsing to enable end-to-end, high-precision conversion of chemical images into standard formats such as SMILES and MolFile. Evaluated on the MolRecBench-Wild dataset, the method achieves a SMILES exact-match accuracy of 93.02%, substantially outperforming the previous state-of-the-art approach (74.87%). The system is now publicly available as part of the MinerU platform.

Abstract

In organic chemistry papers and patents, molecular structures, reaction schemes, and experimental conditions are often presented as molecular structure depictions, reaction diagrams, and complex tables or figures. Such information is difficult for general-purpose document parsing systems to directly convert into machine-readable data. This limits data production for organic chemistry knowledge base construction and for AI for Chemistry tasks such as reaction prediction, retrosynthesis, condition recommendation, molecular property prediction, and drug molecule design. This report introduces MinerU.Chem, a document parsing system for organic chemistry literature integrated into the MinerU online platform. Built on top of MinerU's general document parsing pipeline, MinerU.Chem adds five chemistry-specific modules: chemistry relevance filtering, molecular structure detection, molecule identifier extraction, molecular structure recognition, and reaction scheme parsing. Together, these modules convert organic-chemistry-related image regions in documents into a Molecule Summary List and a Reaction Summary List. For molecular structure recognition, MinerU.Chem uses CARBON (Complex Atomic Representation and Bonding Object Notation) as its core representation. CARBON enables recognition results to preserve both the visual layout of the original image and complex chemical semantics, while supporting the export of standard downstream formats such as MolFile and SMILES. On the SMILES-evaluable subset of MolRecBench-Wild (N=2,392), MinerU.Chem's molecular structure recognition module achieves a SMILES exact-match accuracy of 93.02%, outperforming the best evaluated comparison system, GPT-5.6-Sol (74.87%), by 18.15 percentage points. The system has been integrated into the MinerU online platform and is available at https://mineru.net/OpenSourceTools/Extractor .

Problem

• optical chemical structure recognition
• reaction scheme parsing
• document parsing
• machine-readable chemical data
• chemistry knowledge base

Innovation

• CARBON
• molecular structure recognition
• reaction scheme parsing
• chemistry-specific document parsing
• SMILES accuracy