发表机构
INSA Lyon; CNRS; UCBL; GAUC（里昂国立应用科学学院; 法国国家科学研究中心; 里昂第一大学; GAUC）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出C-Unseen框架，通过LLM的思维链推理识别动态时序知识图谱中与主导叙事有张力的罕见子图并追踪其持续性，实现弱信号检测，性能优于各类基线方法。
AI 中文摘要
弱信号是在重大变化确立之前，预示这些变化的早期、低可见度指标。现有基于关键词频率、主题建模或无类型图拓扑的检测方法，无法捕捉此类信号所呈现的语义与关系结构。本文提出C-Unseen，一种用于动态时序知识图谱（DTKG）中弱信号检测的自解释框架。我们将弱信号定义为在连续TKG快照中扩散的罕见、语义连贯的子图。该框架通过两个模块运行：罕见子图提取器模块中，大语言模型（LLM）通过思维链推理识别出与主导快照叙事内容存在张力的子图；弱信号告警器模块中，这些罕见子图在时间步长上的持续性被追踪，以分离出真正的弱信号。实验结果表明，C-Unseen的性能优于基于关键词、主题和图的基线方法。
英文摘要
Weak signals are early, low-visibility indicators that precede significant changes before those changes become established. Existing detection methods, based on keyword frequency, topic modeling, or untyped graph topology, fail to capture the semantic and relational structure through which such signals manifest. In this paper, we propose C-Unseen, a self-interpretable framework for weak signal detection in Dynamic Temporal Knowledge Graphs (DTKGs). We define a weak signal as a rare, semantically coherent subgraph that proliferates across consecutive TKG snapshots. The framework operates through two modules: a Rare Subgraphs Extractor, in which an LLM identifies subgraphs whose content is in tension with the dominant snapshot narrative via chain-of-thought reasoning, and a Weak Signal Alerter, in which the persistence of these rare subgraphs is tracked across time steps to isolate true weak signals. Experimental results demonstrate that C-Unseen outperforms keyword-, topic-, and graph-based baselines.
CommentsAccepted at the AI4SE 2026 Special Track, held within the WISE 2026 Conference
