Despite tremendous progress of generative models in the natural sciences, their
controllability remains challenging. One fundamentally missing aspect of molecular or …

Generative models have significantly influenced both vision and language domains,
ushering in innovative multimodal applications. Although these achievements have …

Foundation models (FMs) are achieving remarkable successes to realize complex
downstream tasks in domains including natural language and visions. In this paper, we …

Undesired toxicity is a major hindrance to drug discovery and largely responsible for high
attrition rates in early stages. This calls for new, reliable, and interpretable molecular …

Recent work showed that active site rather than full-protein-sequence information improves
predictive performance in kinase-ligand binding affinity prediction. To refine the notion of an …

With the growing amount of chemical data stored digitally, it has become crucial to represent
chemical compounds accurately and consistently. Harmonized representations facilitate the …

Patents show how technology evolves in most scientific fields over time. The best way to use
this valuable knowledge base is to use efficient and effective information retrieval and …

Scientific discovery is the process of successful scientific inquiry that allows us to develop
new technologies, solve practical problems, and aid decision-making. This process is a non …

While the capabilities of generative models heavily improved in different domains (images,
text, graphs, molecules, etc.), their evaluation metrics largely remain based on simplified …