# Discrete Diffusion-Based Model-Level Explanation of Heterogeneous GNNs with Node Features

arXiv: https://arxiv.org/abs/2508.08458 · HTML: https://mlsanity.com/paper/2508.08458/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Flow matching, time series and science, one of 14 directions in Diffusion models.
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 11 Aug 2025.

## Placed in

- Diffusion models / [Flow matching, time series and science](https://mlsanity.com/diffusion/flow-matching-time-series-and-science/)

## Builds on (most cited first, 23 of 23)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Revisiting Deep Learning Models for Tabular Data](https://mlsanity.com/paper/2106.11959/) (2021)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [Heterogeneous Graph Attention Network](https://mlsanity.com/paper/1903.07293/) (2019)
- [TUDataset: A collection of benchmark datasets for learning with graphs](https://mlsanity.com/paper/2007.08663/) (2020)
- [DiGress: Discrete Denoising diffusion for graph generation](https://mlsanity.com/paper/2209.14734/) (2022)
- [Parameterized Explainer for Graph Neural Network](https://mlsanity.com/paper/2011.04573/) (2020)
- [GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders](https://mlsanity.com/paper/1802.03480/) (2018)
- [TabDDPM: Modelling Tabular Data with Diffusion Models](https://mlsanity.com/paper/2209.15421/) (2022)
- [MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding](https://mlsanity.com/paper/2002.01680/) (2020)
- [On Explainability of Graph Neural Networks via Subgraph Explorations](https://mlsanity.com/paper/2102.05152/) (2021)
- [XGNN: Towards Model-Level Explanations of Graph Neural Networks](https://mlsanity.com/paper/2006.02587/) (2020)
- [OpenXAI: Towards a Transparent Evaluation of Model Explanations](https://mlsanity.com/paper/2206.11104/) (2022)
- [Evaluating Explainability for Graph Neural Networks](https://mlsanity.com/paper/2208.09339/) (2022)
- [On Node Features for Graph Neural Networks](https://mlsanity.com/paper/1911.08795/) (2019)
- [EvoLearner: Learning Description Logics with Evolutionary Algorithms](https://mlsanity.com/paper/2111.04879/) (2021)
- [Little Ball of Fur: A Python Library for Graph Sampling](https://mlsanity.com/paper/2006.04311/) (2020)
- [MotifExplainer: a Motif-based Graph Neural Network Explainer](https://mlsanity.com/paper/2202.00519/) (2022)
- [D4Explainer: In-Distribution GNN Explanations via Discrete Denoising Diffusion](https://mlsanity.com/paper/2310.19321/) (2023)
- [Heterogeneous Subgraph Transformer for Fake News Detection](https://mlsanity.com/paper/2404.13192/) (2024)
- [Logical Distillation of Graph Neural Networks](https://mlsanity.com/paper/2406.07126/) (2024)
- [Deep Generation of Heterogeneous Networks](https://mlsanity.com/paper/2206.12336/) (2022)
- [Combining Sub-Symbolic and Symbolic Methods for Explainability](https://mlsanity.com/paper/2112.01844/) (2021)

## Abstract (arXiv)

Many real-world datasets, such as citation networks, social networks, and molecular structures, are naturally represented as heterogeneous graphs, where nodes belong to different types and have additional features. For example, in a citation network, nodes representing "Paper" or "Author" may include attributes like keywords or affiliations. A critical machine learning task on these graphs is node classification, which is useful for applications such as fake news detection, corporate risk assessment, and molecular property prediction. Although Heterogeneous Graph Neural Networks (HGNNs) perform well in these contexts, their predictions remain opaque. Existing post-hoc explanation methods lack support for actual node features beyond one-hot encoding of node type and often fail to generate realistic, faithful explanations. To address these gaps, we propose DiGNNExplainer, a model-level explanation approach that synthesizes heterogeneous graphs with realistic node features via discrete denoising diffusion. In particular, we generate realistic discrete features (e.g., bag-of-words features) using diffusion models within a discrete space, whereas previous approaches are limited to continuous spaces. We evaluate our approach on multiple datasets and show that DiGNNExplainer produces explanations that are realistic and faithful to the model's decision-making, outperforming state-of-the-art methods.