University of Notre Dame

Motif-aware graph diffusion model for controllable molecular generation that adapts to unseen properties by learning a lightweight task embedding.

Graph diffusion transformer for in-context molecular design, adapting to new tasks from a few molecule-property demonstrations without fine-tuning.

Multimodal LLM for inverse molecular design, interleaving text and graph generation with a diffusion transformer and A* retrosynthetic planning.

Discrete graph diffusion transformer generating molecules under several numerical and categorical property targets at once, with no guiding predictor.