GradXKG：一种通用的随用随解释时序知识图谱解释器
GradXKG: A Universal Explain-per-use Temporal Knowledge Graph Explainer
AI总结：
针对现有基于RGCN的时序知识图谱推理模型可解释性不足的问题，提出两阶段梯度驱动的通用解释器GradXKG，可高效生成符合模型逻辑的解释，适配多种相关模型架构。
摘要：
时序知识图谱（TKGs）通过引入时间维度来表示事实随时间的演变规律，已在推理任务中展现出应用潜力。然而，现有的时序知识图谱推理（TKGR）模型因黑箱特性缺乏可解释性。近期研究试图通过定制化的模型架构生成推理路径来解决这一问题，但这类方法泛化性有限，且输出的解释内容较为稀疏。为实现大多数TKGR模型的可解释性，我们提出GradXKG——一种新颖的两阶段基于梯度的方法，用于解释基于关系图卷积网络（RGCN）的TKGR模型。第一阶段是受Grad-CAM启发的RGCN解释器，它以高效的“随用随解释”方式追踪梯度，量化每个节点跨时间步的贡献。第二阶段是集成梯度解释器，它整合RGCN输出的重要性得分，扩展了对各类基于RGCN的TKGR架构的兼容性。两个解释器协同工作，可针对给定预测突出每个时间步中最关键的节点。我们的大量实验表明，通过利用梯度信息，GradXKG能够为大多数基于RGCN的TKGR模型及时提供基于模型逻辑的、有洞察力的解释。这有助于解决现有TKGR模型可解释性不足的问题，并提供了一种适用于多种模型的通用解释方法。
英文摘要：
Temporal knowledge graphs (TKGs) have shown promise for reasoning tasks by incorporating a temporal dimension to represent how facts evolve over time. However, existing TKG reasoning (TKGR) models lack explainability due to their black-box nature. Recent work has attempted to address this through customized model architectures that generate reasoning paths, but these recent approaches have limited generalizability and provide sparse explanatory output. To enable interpretability for most TKGR models, we propose GradXKG, a novel two-stage gradient-based approach for explaining Relational Graph Convolution Network (RGCN)-based TKGR models. First, a Grad-CAM-inspired RGCN explainer tracks gradients to quantify each node's contribution across timesteps in an efficient "explain-per-use" fashion. Second, an integrated gradients explainer consolidates importance scores for RGCN outputs, extending compatibility across diverse TKGR architectures based on RGCN. Together, the two explainers highlight the most critical nodes at each timestep for a given prediction. Our extensive experiments demonstrated that, by leveraging gradient information, GradXKG provides insightful explanations grounded in the model's logic in a timely manner for most RGCN-based TKGR models. This helps address the lack of interpretability in existing TKGR models and provides a universal explanation approach applicable across various models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
