PTCL: Pseudo-Label Temporal Curriculum Learning for Label-Limited Dynamic Graph

2025-04-24

AI Summary

Addressing the practical challenge of scarce historical labels—where only final-timestamp node labels are readily available—in dynamic graph node classification, this paper proposes FLiD, a novel framework. FLiD introduces a temporal-decoupled architecture that explicitly separates structural modeling from temporal evolution modeling. It further designs an exponential decay-weighted temporal curriculum learning strategy to progressively guide the model in learning high-confidence pseudo-labels from unlabeled timestamps. Moreover, FLiD establishes the first unified framework and releases a new benchmark dataset, CoOAG, specifically tailored for this setting. Extensive experiments on real-world applications—including financial risk control and academic collaboration—demonstrate that FLiD consistently outperforms adapted state-of-the-art baselines across multiple dynamic graph benchmarks. To foster reproducibility and community advancement, the authors fully open-source their training and evaluation code.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Dynamic node classification is critical for modeling evolving systems like financial transactions and academic collaborations. In such systems, dynamically capturing node information changes is critical for dynamic node classification, which usually requires all labels at every timestamp. However, it is difficult to collect all dynamic labels in real-world scenarios due to high annotation costs and label uncertainty (e.g., ambiguous or delayed labels in fraud detection). In contrast, final timestamp labels are easier to obtain as they rely on complete temporal patterns and are usually maintained as a unique label for each user in many open platforms, without tracking the history data. To bridge this gap, we propose PTCL(Pseudo-label Temporal Curriculum Learning), a pioneering method addressing label-limited dynamic node classification where only final labels are available. PTCL introduces: (1) a temporal decoupling architecture separating the backbone (learning time-aware representations) and decoder (strictly aligned with final labels), which generate pseudo-labels, and (2) a Temporal Curriculum Learning strategy that prioritizes pseudo-labels closer to the final timestamp by assigning them higher weights using an exponentially decaying function. We contribute a new academic dataset (CoOAG), capturing long-range research interest in dynamic graph. Experiments across real-world scenarios demonstrate PTCL's consistent superiority over other methods adapted to this task. Beyond methodology, we propose a unified framework FLiD (Framework for Label-Limited Dynamic Node Classification), consisting of a complete preparation workflow, training pipeline, and evaluation standards, and supporting various models and datasets. The code can be found at https://github.com/3205914485/FLiD.

Problem

• Addresses label-limited dynamic node classification with final timestamp labels only
• Proposes pseudo-label generation via temporal decoupling architecture and decoder
• Introduces Temporal Curriculum Learning to prioritize reliable pseudo-labels

Innovation

• Temporal decoupling architecture separates backbone and decoder
• Temporal Curriculum Learning weights pseudo-labels exponentially
• Unified framework FLiD supports models and datasets