Finding new drugs is getting harder and harder. One of the hopes of drug discovery is to use machine learning models to predict molecular properties.

Thus, in this paper, we seek to establish a new gold standard for small molecule drug discovery benchmarking, W elQrate .

Neural Information Processing Systems http://nips.cc/

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

Nevertheless, two issues impede the usage of GNNs in real scenarios: (1)insufficient labeled molecules forsupervised training; (2)poorgeneralization capability to new-synthesized molecules.

Welcome back to, TIME's new twice-weekly newsletter about AI. If you're reading this in your browser, why not subscribe to have the next one delivered straight to your inbox? We hear a lot about how AI is accelerating drug discovery. But the number of drugs approved by the FDA has remained constant through the AI revolution, at around 50 per year. "The biggest problem in bringing new medicine to patients hasn't been drug discovery for a long time," says Ben Liu, the founder and CEO of Formation Bio, an AI company working in the biotech space.

Structure-based and ligand-based computational drug design have traditionally relied on disjoint data sources and modeling assumptions, limiting their joint use at scale. In this work, we introduce Contrastive Geometric Learning for Unified Computational Drug Design (ConGLUDe), a single contrastive geometric model that unifies structure- and ligand-based training. ConGLUDe couples a geometric protein encoder that produces whole-protein representations and implicit embeddings of predicted binding sites with a fast ligand encoder, removing the need for pre-defined pockets. By aligning ligands with both global protein representations and multiple candidate binding sites through contrastive learning, ConGLUDe supports ligand-conditioned pocket prediction in addition to virtual screening and target fishing, while being trained jointly on protein-ligand complexes and large-scale bioactivity data. Across diverse benchmarks, ConGLUDe achieves state-of-the-art zero-shot virtual screening performance in settings where no binding pocket information is provided as input, substantially outperforms existing methods on a challenging target fishing task, and demonstrates competitive ligand-conditioned pocket selection. These results highlight the advantages of unified structure-ligand training and position ConGLUDe as a step toward general-purpose foundation models for drug discovery.