Advances in 3D Molecular Generation

The field of 3D molecular generation is moving towards more efficient, flexible, and controllable methods. Recent developments have focused on improving the generation quality and speed of diffusion-based models, as well as exploring alternative approaches such as autoregressive models. Noteworthy papers in this area include one that introduces a training-free framework for evolutionary guidance in diffusion models, enabling rapid and guided exploration of chemical space. Another paper presents a scalable autoregressive model that achieves substantial improvements in generation quality and competitiveness with state-of-the-art diffusion models. A third paper introduces a collaborative constrained graph diffusion model that generates molecules guaranteed to be chemically valid, outperforming state-of-the-art approaches while requiring fewer parameters.

Sources

Evolutionary training-free guidance in diffusion model for 3D multi-objective molecular generation

Scalable Autoregressive 3D Molecule Generation

A collaborative constrained graph diffusion model for the generation of realistic synthetic molecules

Learning Flexible Forward Trajectories for Masked Molecular Diffusion

Built with on top of