GNN在比特币欺诈检测中表现不如特征模型，真实图结构可能反成负担。
When Graph Structure Becomes a Liability: A Critical Re-Evaluation of Graph Neural Networks for Bitcoin Fraud Detection under Temporal Distribution Shift
- 在严格归纳设置下，仅用原始特征的随机森林优于所有GNN模型。
- 训练时接触测试期边导致F1下降39.5点，暴露数据泄露问题。
- 打乱边结构反而性能更好，说明真实图拓扑在时间漂移下可能误导模型。
广泛引用的共识认为GCN、GraphSAGE、GAT和EvolveGCN在Elliptic Bitcoin数据集上优于仅使用特征的基线模型，但该结论未经过无泄露评估协议的严格检验。我们进行了种子匹配的归纳与归纳对比实验，发现这一共识不成立。在严格归纳协议下，仅使用原始特征的随机森林模型取得F1 = 0.821，优于所有评估过的GNN模型，而GraphSAGE仅达F1 = 0.689 ± 0.017。配对受控实验表明，因训练时暴露测试期邻接关系，导致F1下降39.5分。边打乱消融实验显示，随机连接图的表现优于真实交易图，表明在时间分布漂移下，数据集拓扑可能具有误导性。结合GNN嵌入与原始特征的混合模型仅带来微弱增益，仍显著低于仅用特征的基线。我们发布了代码、检查点和严格归纳协议，以支持可复现的无泄露评估。
原文摘要 · Abstract (English)
The consensus that GCN, GraphSAGE, GAT, and EvolveGCN outperform feature-only baselines on the Elliptic Bitcoin Dataset is widely cited but has not been rigorously stress-tested under a leakage-free evaluation protocol. We perform a seed-matched inductive-versus-transductive comparison and find that this consensus does not hold. Under a strictly inductive protocol, Random Forest on raw features achieves F1 = 0.821 and outperforms all evaluated GNNs, while GraphSAGE reaches F1 = 0.689 +/- 0.017. A paired controlled experiment reveals a 39.5-point F1 gap attributable to training-time exposure to test-period adjacency. Additionally, edge-shuffle ablations show that randomly wired graphs outperform the real transaction graph, indicating that the dataset's topology can be misleading under temporal distribution shift. Hybrid models combining GNN embeddings with raw features provide only marginal gains and remain substantially below feature-only baselines. We release code, checkpoints, and a strict-inductive protocol to enable reproducible, leakage-free evaluation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。