/Diffusion-based-Graph-Generative-Methods

The survey on diffusion-based graph genrative methods

MIT LicenseMIT

Diffusion-based Grap Generative Methods

This is the summation of all the methods, datasets, and other survey mentioned in our survey 'Diffusion-based Graph Generative Methods'.

Contents

Molecule generation

De novo molecule design

Methods Paper Code Methods Paper Code
DiGress (ICLR-23) [paper] [code] MiDi (ICLR-23) [paper] [code]
CDGS (NeurIPS-22) [paper] [code] GCDM (ICLR-23) [paper] [code]
EDM (ICML-22) [paper] [code] Wu et al. (NeurIPS-22) [paper] -
MDM (AAAI-23) [paper] [code] DiffLinker [paper] [code]
JODO [paper] [code] SILVR [paper] -
HierDiff (ICML-23) [paper] [code] - - -

Conformation design

Methods Paper Code Methods Paper Code
ConfGF (ICML-21) [paper] [code] DGSM (NeurIPS-21) [paper] -
GeoDiff (ICLR-22) [paper] [code] ColfNet (ICML-22) [paper] -
Torsion Diffusion (NeurIPS-22) [paper] [code] DiffMD (AAAI-23) [paper] -
RINGER [paper] - - - -

De novo ligand design

Methods Paper Code Methods Paper Code
DiffBP [paper] - DiffSBDD [paper] -
TargetDiff (ICLR-23) [paper] - PMDM [paper] -
D3FG [paper] - - - -

Ligand docking

Methods Paper Code
DIFFDOCK (NeurIPS-22) [paper] -
EDM-Dock (JCIM) [paper] [code]
DPL [paper] [code]
NeuralPLexer [paper] -

Protein design

Methods Paper Code Methods Paper Code
DiffAb (NeurIPS-22) [paper] - Anand [paper] -
PROTSEED (ICLR-23) [paper] - ProteinSGM [paper] [code]
SMCDiff (ICLR-23) [paper] [code] GraDe-IF [paper] -
EigenFold [paper] [code] - - -

Motion generation

Motion synthesis

Methods Paper Code Homepage Methods Paper Code Homepage
MotionDiffuse [paper] [code] [homepage] Modiff [paper] - -
Ren et al. (ICASSP-23) [paper] - - FLAME (AAAI-23) [paper] - -
MoFusion (CVPR-23) [paper] - [homepage] MDM (ICLR-23) [paper] [code] [homepage]
MLD (CVPR-23) [paper] [code] [homepage] PriorMDM [paper] [code] [homepage]
Alexanderson et al. (ACM Trans. Graph.) [paper] - - EDGE (CVPR-23) [paper] [code] [homepage]
SceneDiffuser [paper] [code] [homepage] MoDi (CVPR-23) [paper] [code] [homepage]
BiGraphDiff [paper] - - DiffuPose [paper] - -

Motion prediction

Methods Paper Code Homepage
Ahn et al. (ICRA-23) [paper] [code] [homepage]
HumanMAC [paper] [code] [homepage]
TCD (ICRA-23) [paper] [code] -
DiffMotion [paper] - -

Others

Methods Paper Code Homepage Methods Paper Code Homepage
EDP-GNN [paper] - - GSDM [paper] - -
NVDiff [paper] - - DPM-GSP [paper] - -
DiffSTG [paper] - - DIFUSCO [paper] - -
GraphGDP [paper] [code] - HouseDiffusion (CVPR-23) [paper] [code] [homepage]
NAP [paper] - [homepage] EDGE [paper] - -
DruM [paper] - - DDM [Paper] - -
DiffusionNAG [paper] - - TSDiff [paper] - -
GraphArm [paper] - - HGDM [paper] [code] -
Lee et al. [paper] - - SaGess [paper] - -
SLD [paepr] - - Diff-POI [paper] - -
Brain Diffuser [paper] - - Lu et al. [paper] - -

Datasets

Molecule generation

Methods Paper Source Methods Paper Source
Zinc [paper] [source] GEOM-QM9 [paper] [source]
GEOM-Drugs [paper] [source] CrossDocked2020 [paper] [source]
BioLiP [paper] [source] PDBBind [paper] [source]
SAbDab [paper] [source] - - -

Motion generation

Methods Paper Source Methods Paper Source
Human3.6M [paper] [source] HumanEva-I [paper] [source]
HumanAct12 [paper] [source] HumanML3D [paper] [source]
KIT [paper] [source] BABEL [paper] [source]
UESTC [paper] [source] 3DPW [paper] [source]
NTU RGB+D [paper] [source] AIST++ [paper] [source]
TSG [paper] [source] ZeroEGGS [paper] [source]

Other surveys

Paper Url Source
Diffusion Models: A Comprehensive Survey of Methods and Applications [paper] [source]
A Survey on Generative Diffusion Model [paper] [source]
Generative Diffusion Models on Graphs: Methods and Applications [paper] -
A Survey on Graph Diffusion Models: Generative AI in Science for Molecule, Protein and Material [paper] -
Graph-based Molecular Representation Learning [paper] -
Generative Models as an Emerging Paradigm in the Chemical Sciences [paper] -
A Survey on Deep Graph Generation: Methods and Applications (LoG-22) [paper] -
A Survey on Temporal Graph Representation Learning and Generative Modeling [paper] -
Human motion modeling with deep learning: A survey [paper] -
MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design [paper] -