Pith. sign in

REVIEW 13 cited by

Energy-based Out-of-Distribution Detection for Graph Neural Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2302.02914 v2 pith:3OA4OPQU submitted 2023-02-06 cs.LG cs.AIcs.SI

classification cs.LGcs.AIcs.SI
keywords datadetectiongraphlearningshiftsenergygnnsafegraphs
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Learning on graphs, where instance nodes are inter-connected, has become one of the central problems for deep learning, as relational structures are pervasive and induce data inter-dependence which hinders trivial adaptation of existing approaches that assume inputs to be i.i.d.~sampled. However, current models mostly focus on improving testing performance of in-distribution data and largely ignore the potential risk w.r.t. out-of-distribution (OOD) testing samples that may cause negative outcome if the prediction is overconfident on them. In this paper, we investigate the under-explored problem, OOD detection on graph-structured data, and identify a provably effective OOD discriminator based on an energy function directly extracted from graph neural networks trained with standard classification loss. This paves a way for a simple, powerful and efficient OOD detection model for GNN-based learning on graphs, which we call GNNSafe. It also has nice theoretical properties that guarantee an overall distinguishable margin between the detection scores for in-distribution and OOD samples, which, more critically, can be further strengthened by a learning-free energy belief propagation scheme. For comprehensive evaluation, we introduce new benchmark settings that evaluate the model for detecting OOD data from both synthetic and real distribution shifts (cross-domain graph shifts and temporal graph shifts). The results show that GNNSafe achieves up to $17.0\%$ AUROC improvement over state-of-the-arts and it could serve as simple yet strong baselines in such an under-developed area.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Hypergraph Neural Stochastic Diffusion: An SDE Framework for Uncertainty Estimation

    cs.LG 2026-07 conditional novelty 6.5 of 10

    HyperNSD models hypergraph node states as an incidence-aware SDE whose pathwise variability yields competitive uncertainty estimates for OOD and misclassification detection.

  2. Both Topology and Text Matter: Revisiting LLM-guided Out-of-Distribution Detection on Text-attributed Graphs

    cs.LG 2026-02 conditional novelty 6.0 of 10

    LG-Plug mines pseudo-OOD exposures from clustered unlabeled nodes via iterative LLM prompting and regularizes topology-driven graph OOD detectors, cutting FPR95 by ≥7% across six TAG benchmarks.

  3. Subgraph Generation for Generalizing on Out-of-Distribution Links

    cs.LG 2025-07 conditional novelty 6.0 of 10

    FLEX is a generative framework that synthesizes counterfactual subgraphs with a semi-implicit graph VAE and adversarially co-trains a GNN to improve out-of-distribution link prediction.

  4. SODA: Out-of-Distribution Detection in Domain-Shifted Point Clouds via Neighborhood Propagation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A training-free neighborhood score propagation method improves out-of-distribution detection for point clouds under synthetic-to-real domain shift.

  5. An Information-Theoretic Framework for Feature Construction in Out-of-Distribution Detection

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A variational loss combining KL divergence and the Information Bottleneck predicts a piecewise-linear shaping function for OOD detection that beats existing element-wise shaping methods on ImageNet and CIFAR benchmarks.

  6. Evidential Spectrum-Aware Contrastive Learning for OOD Detection in Dynamic Graphs

    cs.LG 2025-06 conditional novelty 6.0 of 10

    EviSEC combines evidential uncertainty and spectrum-aware contrastive training to detect out-of-distribution dynamic graphs, reporting strong AUROC gains over prior baselines.

  7. When LLMs meet open-world graph learning: a new perspective for unlabeled data uncertainty

    cs.LG 2025-05 reject novelty 6.0 of 10

    OGA combines prototype-based unknown-class rejection with LLM-generated, structure-guided annotations so text-attributed graphs can be retrained in open-world settings.

  8. Graph Synthetic Out-of-Distribution Exposure with Large Language Models

    cs.LG 2025-04 conditional novelty 6.0 of 10

    LLM-identified or LLM-generated pseudo-OOD nodes used as exposure data during GNN training improve node-level OOD detection on text-attributed graphs without real OOD labels.

  9. GLIP-OOD: Zero-Shot Graph OOD Detection with Graph Foundation Model

    cs.LG 2025-04 conditional novelty 6.0 of 10

    GLIP-OOD performs zero-shot node-level OOD detection on text-attributed graphs using a graph foundation model, optionally with LLM-generated pseudo-OOD labels, and matches or exceeds supervised baselines.

  10. Towards Precise Prediction Uncertainty in GNNs: Refining GNNs with Topology-grouping Strategy

    cs.LG 2024-12 conditional novelty 6.0 of 10

    A post-hoc GNN calibration method that groups nodes by confidence and neighborhood prediction similarity and applies group-specific temperatures, reducing expected calibration error on many benchmarks.

  11. Bounded and Uniform Energy-based Out-of-distribution Detection for Graphs

    cs.LG 2025-04 conditional novelty 5.0 of 10

    NODESAFE adds two logit variance penalties to GNN training, cutting FPR95 for node-level OOD detection dramatically compared with GNNSAFE on citation and social-network graphs.

  12. SpectralGap: Graph-Level Out-of-Distribution Detection via Laplacian Eigenvalue Gaps

    cs.LG 2025-05 conditional novelty 4.0 of 10

    SpecGap adjusts GNN features by subtracting the second-largest Laplacian eigenvector component times the spectral gap, claiming improved graph OOD detection without retraining.

  13. Handling Out-of-Distribution Data: A Survey

    cs.LG 2025-07 conditional novelty 3.0 of 10

    A survey that organizes covariate and semantic shift handling methods into one taxonomy and argues for unified models, while contributing no new experiments or benchmark evaluation.

Pith tools