Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
Molecule Transformers provides a collection of recipes for pre-training and fine-tuning molecular transformer language models like BART and BERT for drug discovery. The project introduces novel SMILES pre-training objectives based on contrastive learning and denoising to improve performance in low-data settings.
Parse Score