MolNexTR proposes a dual-stream architecture combining ConvNext and Vision Transformers to improve molecular image recognition (OCSR). It achieves 81-97% accuracy across diverse benchmarks utilizing simultaneous local and global feature extraction alongside specialized image contamination augmentations.
The MolParser project introduces two key datasets: MolParser-7M, the largest training dataset for Optical Chemical Structure Recognition (OCSR) with 7.7M pairs of images and E-SMILES strings, and WildMol, a new 20k-sample benchmark for evaluating models on challenging real-world data. The training data uniquely combines millions of diverse synthetic molecules with 400,000 manually annotated in-the-wild samples.
A 2025 end-to-end OCSR system addressing both technical and data challenges, introducing MolParser-7M (7M+ image-text pairs) and MolDet (YOLO-based detector) for extracting and recognizing molecular structures from real-world documents with diverse quality and styles.
ZINC-22 is a multi-billion-scale public database containing over 37 billion make-on-demand molecules. It utilizes distributed infrastructure and specialized search algorithms to support modern ultra-large virtual screening campaigns.
Build a robust Python CLI tool that converts both SMILES and SELFIES notation into publication-quality 2D molecular images, complete with formulas and legends.
SELFIES is a molecular string representation where every possible string decodes to a valid molecule, solving the invalid-output problem that limits SMILES in generative machine learning.
MARCEL provides a comprehensive benchmark for molecular representation learning with 722K+ conformers across four diverse subsets (Drugs-75K, Kraken, EE, BDE), enabling evaluation of conformer ensemble methods for property prediction in drug discovery and catalysis.
Comprehensive overview of SMILES notation for chemical structures, covering syntax for atoms, bonds, branches, rings, and stereochemistry, plus its key limitations for machine learning.
A foundational 1931 paper that derives exact recursive formulas for counting alkane structural isomers, correcting historical errors and establishing the first systematic enumeration up to C₄₀.
GEOM contains 450k+ molecules with 37M+ conformations, featuring energy annotations from semi-empirical (GFN2-xTB) and DFT methods for property prediction and molecular generation research.
GDB-11 contains 26.4 million systematically generated small organic molecules with up to 11 atoms, establishing the methodology for exploring drug-like chemical space computationally.
ICML 2025 methodological paper introducing SPHNet, which uses adaptive network sparsification to overcome the computational bottleneck of tensor products in SE(3)-equivariant networks, achieving up to 7x speedup and 75% memory reduction on DFT Hamiltonian prediction tasks.