Abstract:Denoising diffusion models have shown great potential in multiple research areas. Existing diffusion-based generative methods on de novo 3D molecule generation face two major challenges. Since majority heavy atoms in molecules allow connections to multiple atoms through single bonds, solely using pair-wise distance to model molecule geometries is insufficient. Therefore, the first one involves proposing an effective neural network as the denoising kernel that is capable to capture complex multi-body interatomic relationships and learn high-quality features. Due to the discrete nature of graphs, mainstream diffusion-based methods for molecules heavily rely on predefined rules and generate edges in an indirect manner. The second challenge involves accommodating molecule generation to diffusion and accurately predicting the existence of bonds. In our research, we view the iterative way of updating molecule conformations in diffusion process is consistent with molecular dynamics and introduce a novel molecule generation method named Geometric-Facilitated Molecular Diffusion (GFMDiff). For the first challenge, we introduce a Dual-Track Transformer Network (DTN) to fully excevate global spatial relationships and learn high quality representations which contribute to accurate predictions of features and geometries. As for the second challenge, we design Geometric-Facilitated Loss (GFLoss) which intervenes the formation of bonds during the training period, instead of directly embedding edges into the latent space. Comprehensive experiments on current benchmarks demonstrate the superiority of GFMDiff.

DiGress: Discrete Denoising diffusion for graph generation

SaGess: Sampling Graph Denoising Diffusion Model for Scalable Graph Generation

Sparse Training of Discrete Diffusion Models for Graph Generation

Autoregressive Diffusion Model for Graph Generation

Graph Generation with Diffusion Mixture

Discrete-state Continuous-time Diffusion for Graph Generation

Cometh: A continuous-time discrete-state graph diffusion model

Conditional Diffusion Based on Discrete Graph Structures for Molecular Graph Generation

Graph Diffusion Transformers for Multi-Conditional Molecular Generation

Graph-Based Generative Adversarial Networks for Molecular Generation with Noise Diffusion.

Diffusion Models for Graphs Benefit From Discrete State Spaces

Pard: Permutation-Invariant Autoregressive Diffusion for Graph Generation

DeFoG: Discrete Flow Matching for Graph Generation

Advancing Graph Generation through Beta Diffusion

NetDiff: Deep Graph Denoising Diffusion for Ad Hoc Network Topology Generation

Graphusion: Latent Diffusion for Graph Generation

GraphGDP: Generative Diffusion Processes for Permutation Invariant Graph Generation

CDGraph: Dual Conditional Social Graph Synthesizing via Diffusion Model

Generative Diffusion Models on Graphs: Methods and Applications

SubGDiff: A Subgraph Diffusion Model to Improve Molecular Representation Learning

Geometric-Facilitated Denoising Diffusion Model for 3D Molecule Generation