Molecular-Representation - Page 2

Multi-modal transformer combining vision, text, and layout encoding to extract complex Markush structures from patent …

A three-stage OCSR framework using SMILES pretraining, auxiliary bond/coordinate tasks, and reinforcement learning to …

Novel Ring-Free Language representation and Molecular Skeleton Decoder architecture for improved optical chemical …

Deep learning OCSR model using keypoint estimation to detect atom and bond centers for graph-based molecular structure …

Deep learning framework using CNN-LSTM image captioning to convert hand-drawn hydrocarbon structures into SMILES strings …

Transformer-based approach for Optical Chemical Structure Recognition converting chemical images to SELFIES strings with …

Vision Transformer encoder with Transformer decoder for molecular image-to-InChI translation, achieving state-of-the-art …

A Transformer-based model (ICMDT) for converting chemical structure images into InChI text strings using a novel Deep …

A deep learning model that converts molecular images directly into graph structures, enabling recognition of abbreviated …

Transformer-based OCSR using a novel synthetic data generation pipeline for robust molecular image interpretation across …

Encoder-decoder model using pre-trained ResNet and attention-based LSTM to translate molecular images into SMILES, …

Ablation study comparing SMILES, DeepSMILES, SELFIES, and InChI for OCSR. SMILES achieves highest accuracy; SELFIES …