Sparse training of discrete diffusion models for graph generation

[PDF] arxiv.org

Generative flows on discrete state-spaces: Enabling multimodal flows with applications to protein co-design

A Campbell, J Yim, R Barzilay, T Rainforth… - arXiv preprint arXiv …, 2024 - arxiv.org
Combining discrete and continuous data is an important capability for generative models.
We present Discrete Flow Models (DFMs), a new flow-based model of discrete data that …
被引用次数：57 相关文章 所有 6 个版本
[PDF] arxiv.org

Generative modelling of structurally constrained graphs

M Madeira, C Vignac, D Thanou, P Frossard - arXiv preprint arXiv …, 2024 - arxiv.org
Graph diffusion models have emerged as state-of-the-art techniques in graph generation,
yet integrating domain knowledge into these models remains challenging. Domain …
被引用次数：1 相关文章 所有 4 个版本
[PDF] arxiv.org

Unlocking Guidance for Discrete State-Space Diffusion and Flow Models

H Nisonoff, J Xiong, S Allenspach… - arXiv preprint arXiv …, 2024 - arxiv.org
Generative models on discrete state-spaces have a wide range of potential applications,
particularly in the domain of natural sciences. In continuous state-spaces, controllable and …
被引用次数：8 相关文章 所有 2 个版本
[PDF] arxiv.org

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

A Siraudin, FD Malliaros, C Morris - arXiv preprint arXiv:2406.06449, 2024 - arxiv.org
Discrete-state denoising diffusion models led to state-of-the-art performance in graph
generation, especially in the molecular domain. Recently, they have been transposed to …
被引用次数：1 相关文章 所有 2 个版本
[PDF] arxiv.org

How Discrete and Continuous Diffusion Meet: Comprehensive Analysis of Discrete Diffusion Models via a Stochastic Integral Framework

Y Ren, H Chen, GM Rotskoff, L Ying - arXiv preprint arXiv:2410.03601, 2024 - arxiv.org
Discrete diffusion models have gained increasing attention for their ability to model complex
distributions with tractable sampling and inference. However, the error analysis for discrete …
相关文章 所有 4 个版本
[PDF] arxiv.org

GraphRCG: Self-conditioned Graph Generation via Bootstrapped Representations

S Wang, Z Tan, X Zhao, T Chen, H Liu, J Li - arXiv preprint arXiv …, 2024 - arxiv.org
Graph generation generally aims to create new graphs that closely align with a specific
graph distribution. Existing works often implicitly capture this distribution through the …
被引用次数：1 相关文章 所有 3 个版本
[PDF] arxiv.org

Heterogeneous Graph Generation: A Hierarchical Approach using Node Feature Pooling

H Ghosh, C Changyu, A Sinha, S Sural - arXiv preprint arXiv:2410.11972, 2024 - arxiv.org
Heterogeneous graphs are present in various domains, such as social networks,
recommendation systems, and biological networks. Unlike homogeneous graphs …
相关文章 所有 2 个版本
[PDF] openreview.net

GLAD: Improving Latent Graph Generative Modeling with Simple Quantization

Y Boget, F Lavda, A Kalousis - ICML 2024 Workshop on Structured … - openreview.net
Exploring the graph latent structures has not garnered much attention in the graph
generative research field. Yet, exploiting the latent space is as crucial as working on the data …
相关文章