提出量化图神经网络长程交互能力的新方法，助力评估模型真实长程建模能力。
On Measuring Long-Range Interactions in Graph Neural Networks
- 定义图算子的长程作用范围度量，从理论层面刻画长程交互
- 通过合成实验验证该度量的有效性，发现多数任务实际长程性不足
- 为新架构和数据集评估提供可复现的分析工具，适合研究者参考
长程图任务——依赖远距离节点间交互的任务——是图神经网络研究中的开放问题。当前主流的长期图基准（Long Range Graph Benchmark）虽被广泛用于验证模型的长程能力，但其评估方式仅基于经验，缺乏鲁棒性和理论支撑；亟需一种更严谨的长程问题表征。为此，本文形式化了图任务中的长程交互，并引入一种针对图算子的范围度量，通过合成实验进行验证。随后，利用该度量分析常见任务与架构，探讨它们实际是否具备长程特性。我们相信本工作推进了对图上长程问题的定义与应对，所提出的范围度量将有助于新数据集与架构的评估。
原文摘要 · Abstract (English)
Long-range graph tasks -- those dependent on interactions between distant nodes -- are an open problem in graph neural network research. Real-world benchmark tasks, especially the Long Range Graph Benchmark, have become popular for validating the long-range capability of proposed architectures. However, this is an empirical approach that lacks both robustness and theoretical underpinning; a more principled characterization of the long-range problem is required. To bridge this gap, we formalize long-range interactions in graph tasks, introduce a range measure for operators on graphs, and validate it with synthetic experiments. We then leverage our measure to examine commonly used tasks and architectures, and discuss to what extent they are, in fact, long-range. We believe our work advances efforts to define and address the long-range problem on graphs, and that our range measure will aid evaluation of new datasets and architectures.
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。