MinerU.Chem system achieves 93% accuracy in chemical structure recognition

A new system called MinerU.Chem has been developed to extract chemical structures and reactions from organic chemistry documents, converting them into machine-readable data. This system, integrated into the MinerU platform, uses a core representation called CARBON to preserve visual layout and chemical semantics. In evaluations, MinerU.Chem achieved a 93.02% accuracy in recognizing molecular structures, significantly outperforming GPT 5.6 "Sol" by over 18 percentage points on the MolRecBench-Wild dataset. AI

IMPACT This system could accelerate AI for Chemistry tasks like reaction prediction and drug design by enabling better data extraction from scientific literature.

RANK_REASON The item is a research paper detailing a new system for chemical structure recognition. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• CARBON
• GPT 5.6 "Sol"
• MDL Molfile
• MinerU.Chem
• MolRecBench-Wild
• Smiles

• paper
• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →