mlsanityData to 7 Oct 2026

Unified Generative Modeling of 3D Molecules via Bayesian Flow Networks

arXiv 2403.15441 ↗First posted 17 Mar 2024physics.chem-ph cs.AI cs.LGcited by 18

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Advanced generative model (e.g., diffusion model) derived from simplified continuity assumptions of data distribution, though showing promising progress, has been difficult to apply directly to geometry generation applications due to the multi-modality and noise-sensitive nature of molecule geometry. This work introduces Geometric Bayesian Flow Networks (GeoBFN), which naturally fits molecule geometry by modeling diverse modalities in the differentiable parameter space of distributions. GeoBFN maintains the SE-(3) invariant density modeling property by incorporating equivariant inter-dependency modeling on parameters of distributions and unifying the probabilistic modeling of different modalities. Through optimized training and sampling techniques, we demonstrate that GeoBFN achieves state-of-the-art performance on multiple 3D molecule generation benchmarks in terms of generation quality (90.87% molecule stability in QM9 and 85.6% atom stability in GEOM-DRUG. GeoBFN can also conduct sampling with any number of steps to reach an optimal trade-off between efficiency and quality (e.g., 20-times speedup without sacrificing performance).

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 24 of its 33 references that are in the map. First posted on arXiv 17 Mar 2024.

Builds on

24 of 33 references.

Closest ancestors · 6

Same direction, newest first

  1. 2023Bayesian Flow Networks14 Aug
  2. 2023Geometric Latent Diffusion Models for 3D Molecule Generation2 May
  3. 2022Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem8 Jun
  4. 2022Protein Structure and Sequence Generation with Equivariant Denoising Diffusion Probabilistic Models26 May
  5. 2022Equivariant Diffusion for Molecule Generation in 3D31 Mar
  6. 2022GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation6 Mar

Other references · 18

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,903
  3. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,006
  4. 2021E(n) Equivariant Graph Neural Networkscited by 951
  5. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  6. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 308
  7. 2020GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generationcited by 301
  8. 2019Cormorant: Covariant Molecular Neural Networkscited by 262
  9. 2020Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densitiescited by 248
  10. 2018Constrained Graph Variational Autoencoders for Molecule Designcited by 182
  11. 2022Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pocketscited by 182
  12. 2019Symmetry-adapted generation of 3d point sets for the targeted discovery of moleculescited by 171
6 more
  1. 2022Diffusion-based Molecule Generation with Informative Prior Bridgescited by 104
  2. 2021Inverse design of 3d molecular structures with conditional generative neural networkscited by 93
  3. 2023MolDiff: Addressing the Atom-Bond Inconsistency Problem in 3D Molecule Diffusion Generationcited by 65
  4. 2022DiffBP: Generative Diffusion of 3D Molecules for Target Protein Bindingcited by 58
  5. 2021Learning to design drug-like molecules in three-dimensional space using deep generative modelscited by 37
  6. 2020Generating 3D Molecular Structures Conditional on a Receptor Binding Site with Deep Generative Modelscited by 33

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 8; 2025: 6; 2026: 4.

18 papers.

Sort:

By area:

  1. 2024MolCRAFT: Structure-Based Drug Design in Continuous Parameter Spacecited by 48
  2. 2024CBGBench: Fill in the Blank of Protein-Molecule Complex Binding Graphcited by 19
  3. 2024Geometric Representation Condition Improves Equivariant Molecule Generationcited by 13
  4. 2025Applications of Modular Co-Design for De Novo 3D Molecule Generationcited by 13
  5. 2025Scalable Autoregressive 3D Molecule Generationcited by 7
  6. 2024Empower Structure-Based Molecule Optimization with Gradient Guided Bayesian Flow Networkscited by 6
  7. 2025Piloting Structure-Based Drug Design via Modality-Specific Optimal Schedulecited by 6
  8. 2025Guiding Diffusion Models with Reinforcement Learning for Stable Molecule Generationcited by 4
  9. 2026Elign: Equivariant Diffusion Model Alignment from Foundational Machine Learning Force Fieldscited by 3
  10. 2026RMFlow: Refined Mean Flow by a Noise-Injection Step for Multimodal Generationcited by 3
  11. 2024ParamReL: Learning Parameter Space Representation via Progressively Encoding Bayesian Flow Networkscited by 2
  12. 2024Any-Property-Conditional Molecule Generation with Self-Criticism using Spanning Treescited by 2
6 more
  1. 2025NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generationcited by 2
  2. 2024Tokenizing 3D Molecule Structure with Quantized Spherical Coordinatescited by 1
  3. 2025PhysicsGen: Can Generative Models Learn from Images to Predict Complex Physical Relations?cited by 1
  4. 2024Marked Temporal Bayesian Flow Point Processescited by 0
  5. 2026A Systematic Evaluation of Molecule Generation Models for De Novo Drug Design: From Benchmarks to Practical Insightscited by 0
  6. 2026TreeRef-BFN: Equivariance-Free De Novo Molecule Generation based on 2D Topology and Internal 3D Geometrycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Geometry-Complete Diffusion for 3D Molecule Generation and Optimization
  2. 2023Equivariant Flow Matching with Hybrid Probability Transport
  3. 2022MDM: Molecular Diffusion Model for 3D Molecule Generation
  4. 2026Geometric Flow Matching for Molecular Conformation Generation via Manifold Decomposition
  5. 2025FlowMol3: Flow Matching for 3D De Novo Small-Molecule Generation
  6. 2025Controllable 3D Molecular Generation for Structure-Based Drug Design Through Bayesian Flow Networks and Gradient Integration
  7. 2024Mixed Continuous and Categorical Flow Matching for 3D De Novo Molecule Generation
  8. 2024A survey of probabilistic generative frameworks for molecular simulations