New models improve chemical structure recognition from images

Researchers have developed two new models, OCSRGlyph and MarkushGlyph, to improve the recognition of chemical structures from images. OCSRGlyph enhances the translation of single molecule images into line notations by better accounting for stereochemistry. MarkushGlyph, a vision-language model, addresses the more complex task of parsing Markush structures, which represent families of molecules, by processing the entire structure as an image in a single stage. The work also introduces a new metric for evaluating Markush structure translations. AI

IMPACT These models could streamline the process of indexing chemical literature and constructing machine learning training sets.

RANK_REASON The cluster contains a research paper detailing new models for chemical structure recognition. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• MarkushGlyph
• OCSRGlyph
• ScienceCast
• Siddharth Narayanan

• paper
• other

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