MarkushGrapher-2: End-to-end Multimodal Recognition of Chemical Structures

Tim Strohmeyer, Lucas Morin, Gerhard Ingmar Meijer, Valéry Weber, Ahmed Nassar, Peter Staar · 2026

Read MarkushGrapher-2: End-to-end Multimodal Recognition of Chemical Structures by Tim Strohmeyer, Lucas Morin, Gerhard Ingmar Meijer, Valéry Weber, Ahmed Nassar, Peter Staar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-28550.txt.

Description

Strohmeyer and co-authors present MarkushGrapher-2, a system that reads chemical structures from documents. Markush structures mix image and text cues, and older tools read them poorly. Their method first uses a dedicated OCR model on chemical images. It then joins text, image, and layout data with two encoders. A two-stage training plan fuses those signals to write the structure. The team also built a large real-world dataset and the IP5-M benchmark. Tests beat prior top models on Markush reading and stayed strong on plain molecules.

Catalog record

CREATORTim Strohmeyer, Lucas Morin, Gerhard Ingmar Meijer, Valéry Weber, Ahmed Nassar, Peter Staar
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE42421
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.28550

Browse the full library · About the Tower of Babel