p–LSGR: A lightweight post–hoc latent–space resampling framework for imbalanced and complex graph distributions

Existing imbalance-mitigation strategies provide only partial robustness, are often computationally expensive, and leave post-hoc imbalance handling largely unexplored. We propose Post-hoc Latent Space Graph Resampling (p-LSGR), a framework that exploits learned graph representations through latent-space resampling to improve minority-class learning without graph augmentation or modification of the original graph structure.

Through data-aware, difficulty-aware, and ensemble-based resampling strategies, p–LSGR improves minority-class classification performance over vanilla GNNs while remaining competitive with augmentation-based approaches, with the ensemble-based variants demonstrating greater stability across encoder types. Across diverse benchmarks, p-LSGR achieves an average 42% improvement in minority-class F1 over vanilla GNNs and a 6% improvement over competing approaches while reducing processing time by up to 91%.