Research map / Graph learning / Graph self-supervised learning
Diverse and Sparse Mixture-of-Experts for Causal Subgraph–Based Out-of-Distribution Graph Learning
Jerry Sun, Mohamed Abubakr Hassan, Yaoyu Zhang, Wanying Zhang, Chi-Guhn Lee
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 diversity loss among experts (ablation) | DiSCO · GOOD-Motif Size | accuracy: 70.07 -> 75.52 | Improved |
| add diversity loss among experts (ablation) | DiSCO · GOOD-HIV Scaffold | ROC-AUC: 65.95 -> 71.55 | Improved |
| add gate loss (teacher-student, sparsity, balance) (ablation) | DiSCO · GOOD-HIV Size | ROC-AUC: 61.91 -> 66.98 | Improved |
| add keep-rate regularization loss (ablation) | DiSCO · GOOD-Motif Basis | accuracy: 67.48 -> 92.80 | Improved |
| add keep-rate regularization loss (ablation) | DiSCO · GOOD-SST2 Length | accuracy: 83.46 -> 83.73 | No clear effect |
| add diversity loss (ablation) | DiSCO · GOOD-Twitter Length | accuracy: 60.10 -> 61.13 | No clear effect |
| add semantic diversity objective (ablation) | DiSCO · Twitter, SST2, Motif-Basis | per-batch loss gap: Twitter 0.13 -> 0.19; SST2 0.07 -> 0.22; Motif-Basis 0.076 -> 0.12 | Improved |
| increase experts from 1 to 4 (ablation) | DiSCO · GOOD benchmark (6 splits, avg) | average accuracy/ROC-AUC: 70.84 -> 74.68 | Improved |
| increase experts from 4 to 8 (ablation) | DiSCO · GOOD benchmark (6 splits, avg) | average accuracy/ROC-AUC: 74.68 -> 75.29 | Mixed |
| perturb keep-rate prior by +/-0.1 or 0.2 (ablation) | DiSCO · Twitter, SST2, Motif-Basis | accuracy: Motif-Basis 92.80 -> 90.13 at -0.2 | No clear effect |
| use single-expert causal subgraph extractor with only sparsity prior | DiSCO (1 expert) · GOOD benchmark (6 splits, avg) | average accuracy/ROC-AUC: 64.24 (ERM) -> 70.84 | Improved |
| use MoE of diverse causal-subgraph experts instead of single-invariant-subgraph methods | DiSCO vs LECI · CFP-Motif (covariate, FIIF, PIIF) | accuracy: 83.20 -> 90.83; 77.73 -> 84.17; 69.40 -> 77.19 | Improved |
Closest papers by meaning
- Quantifying Distributional Invariance in Causal Subgraph for IRM-Free Graph Generalization (NeurIPS 2025 · similarity 0.88)
- Subgraph Aggregation for Out-of-Distribution Generalization on Graphs (AAAI 2025 · similarity 0.88)
- Out-of-Distribution Generalization on Graphs via Progressive Inference (AAAI 2025 · similarity 0.88)
- Improving Out-of-Distribution Generalization in Graphs via Hierarchical Semantic Environments (CVPR 2024 · similarity 0.87)
- A Recipe for Causal Graph Regression: Confounding Effects Revisited (ICML 2025 · similarity 0.87)
- Graph Sparsification via Mixture of Graphs (ICLR 2025 · similarity 0.87)
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.
