MolMole: Molecule Mining from Scientific Literature

2025-04-30

AI Summary

Automated extraction of molecular structures and reaction data from scientific literature is hindered by chemical representation diversity, document unstructuredness, and complex layouts. This paper introduces the first end-to-end vision-driven framework that jointly performs molecule instance detection, reaction graph topology parsing, and optical chemical structure recognition (OCSR) directly from full-page PDFs or images. We unify these three tasks within a single model architecture, construct the first page-level benchmark dataset comprising 550 annotated pages with MOLfile ground truth, and propose dedicated evaluation metrics. Our method integrates document layout-aware detection with reaction graph structural reasoning. It achieves state-of-the-art performance on our proprietary benchmark and multiple public datasets. The code, pre-trained models, and an interactive online demo will be publicly released.

Technology Category

• Computer Vision: Representation Learning for Vision
• Machine Learning: Structured Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

The extraction of molecular structures and reaction data from scientific documents is challenging due to their varied, unstructured chemical formats and complex document layouts. To address this, we introduce MolMole, a vision-based deep learning framework that unifies molecule detection, reaction diagram parsing, and optical chemical structure recognition (OCSR) into a single pipeline for automating the extraction of chemical data directly from page-level documents. Recognizing the lack of a standard page-level benchmark and evaluation metric, we also present a testset of 550 pages annotated with molecule bounding boxes, reaction labels, and MOLfiles, along with a novel evaluation metric. Experimental results demonstrate that MolMole outperforms existing toolkits on both our benchmark and public datasets. The benchmark testset will be publicly available, and the MolMole toolkit will be accessible soon through an interactive demo on the LG AI Research website. For commercial inquiries, please contact us at href{mailto:[e-mail]}{[e-mail]}.

Problem

• Extracting molecular structures from unstructured scientific documents
• Parsing reaction diagrams and chemical data automatically
• Lacking standard benchmarks for page-level chemical data extraction

Innovation

• Vision-based deep learning for molecule detection
• Unified pipeline for chemical data extraction
• Novel benchmark and evaluation metric