Research map / Graph learning / Graph self-supervised learning
Generating Directed Graphs with Dual Attention and Asymmetric Encoding
Alba Carballo-Castro, Manuel Madeira, Yiming QIN, Dorina Thanou, Pascal Frossard
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add dual attention instead of doubling network depth (RRWP) (ablation) | Directo · ER-DAG | V.U.N. (%): 72.0 -> 94.0 | Improved |
| add dual attention instead of doubling network depth (MagLap) (ablation) | Directo · ER-DAG | V.U.N. (%): 80.0 -> 91.0 | Improved |
| add dual attention instead of doubling network depth (RRWP) (ablation) | Directo · SBM | V.U.N. (%): 0.0 -> 99.5 | Improved |
| add dual attention instead of doubling network depth (MagLap) (ablation) | Directo · SBM | V.U.N. (%): 8.0 -> 77.0 | Improved |
| add dual attention instead of doubling network depth (RRWP) (ablation) | Directo · SBM | graph statistics ratio: 26.3 -> 1.8 | Improved |
| add dual attention instead of doubling network depth (RRWP) (ablation) | Directo · ER-DAG | graph statistics ratio: 4.9 -> 1.7 | Improved |
| add positional encodings (ablation) | Directo · ER-DAG and SBM | V.U.N. and Ratio | Improved |
| use direction-aware positional encodings instead of direction-agnostic Laplacian (ablation) | Directo · ER-DAG and SBM | V.U.N. and Ratio | Improved |
| use dual attention with no positional encoding (ablation) | Directo · ER-DAG and SBM | V.U.N. | Improved |
| use Directo flow matching with RRWP instead of DeFoG | Directo RRWP · Visual Genome | V.U.N. (%): 50.8 -> 83.8 | Improved |
| use Directo MagLap instead of LayerDAG | Directo MagLap · TPU Tiles | graph statistics ratio: 413.6 -> 44.0 | Improved |
| increase training data amount (ablation) | Directo · synthetic | generation performance | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- DeFoG — DeFoG: Discrete Flow Matching for Graph Generation (ICML 2025) · better on 6 of 6 shared results
- DiGress — DiGress: Discrete Denoising diffusion for graph generation (ICLR 2023) · better on 6 of 6 shared results
- LayerDAG — LayerDAG: A Layerwise Autoregressive Diffusion Model for Directed Acyclic Graph Generation (ICLR 2025) · better on 3 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- On the Role of Edge Dependency in Graph Generative Models (ICML 2024 · similarity 0.88)
- DeFoG: Discrete Flow Matching for Graph Generation (ICML 2025 · similarity 0.87)
- Smooth Interpolation for Improved Discrete Graph Generative Models (ICML 2025 · similarity 0.87)
- Geometric Generative Modeling with Noise-Conditioned Graph Networks (ICML 2025 · similarity 0.87)
- A Deep Probabilistic Framework for Continuous Time Dynamic Graph Generation (AAAI 2025 · similarity 0.87)
- Efficient and Scalable Graph Generation through Iterative Local Expansion (ICLR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
