Notably, in the blind docking setting, DeltaDock achieves a 31% relative improvement over the docking success rate compared with the previous state-of-the-art GDL model DiffDock.

However, these methods have notable limitations.

Following this thought, we recast virtual screening as an information retrieval task, i.e., given a

However, most existing machine learning methods for small molecule generation suffer from poor synthesizability of candidate compounds, making experimental validation difficult.

Protein-ligand binding prediction is a fundamental problem in AI-driven drug discovery.

However, they depend on scarce experimental data or time-consuming docking simulation, leading to overfitting issues with limited training data and slow generation speed.