发表机构
Southwest Minzu University; University of Electronic Science and Technology of China; Zhejiang University; Tencent（西南民族大学; 电子科技大学; 浙江大学; 腾讯）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有扩散式时序知识图谱外推方法的目标判别信号削弱问题，提出FreqDiff频率感知扩散框架，通过双流去噪器与频域正则化项提升性能，在四个公开基准上达最优表现。
AI 中文摘要
时序知识图谱（Temporal Knowledge Graph, TKG）外推旨在从随时间变化的关系历史中推断未来事实。近期基于扩散的方法通过生成式去噪改进了不确定性建模，但它们对主体历史的聚合条件可能无法充分区分查询特定证据与非显著历史事实，从而削弱目标判别信号。为弥合这一差距，我们提出FreqDiff，一种用于TKG外推的频率感知扩散框架。具体而言，FreqDiff将未来对象预测表述为查询槽去噪，并开发了双流去噪器，其将时间依赖建模与上下文感知谱校准相结合。谱分支从可学习基合成历史条件滤波器，以自适应地重新校准去噪表示；同时提出频域正则化项，以在谱空间中将去噪后的目标与真实对象对齐。在四个公开TKG基准上的实验表明，FreqDiff实现了最先进的性能。
英文摘要
Temporal Knowledge Graph (TKG) extrapolation seeks to infer future facts from time-varying relational histories. Recent diffusion-based approaches improve uncertainty modeling through generative denoising, but their aggregated conditioning on subject histories may insufficiently distinguish query-specific evidence from non-salient historical facts, thereby diluting target-discriminative signals. To bridge this gap, we propose FreqDiff, a Frequency-aware Diffusion framework for TKG extrapolation. Specifically, FreqDiff formulates future object prediction as query-slot denoising and develops a dual-stream denoiser that integrates temporal dependency modeling with context-aware spectral calibration. The spectral branch synthesizes history-conditioned filters from learnable bases to adaptively re-calibrate denoising representations, while a frequency-domain regularizer is proposed to align the denoised target with the gold object in spectral space. Experiments on four public TKG benchmarks demonstrate that FreqDiff achieves state-of-the-art performance.
CommentsEMNLP 2026 Main
