Lune

ACL Findings2026

LANTERN in the Event Stream: Training-Free Temporal Knowledge Graph Forecasting by Balancing Inertia and Shifts

Chengyuan Jin, Ao Chang, Daojian Zeng, Wenhao Teng, Xiangwen Liao, Kang Liu, Jun Zhao, Yubo Chen

摘要

Temporal knowledge graph forecasting (TKGF) asks a model to rank the most plausible future entity for a query such as (s, r, ?, t) from historical events. Recent training-free methods use large language models (LLMs) for this task, but their accuracy depends heavily on which past events are shown in the prompt under a tight context budget. We present LANTERN, a training-free prompting framework that addresses this bottleneck by combining two complementary views of history: a long-window strength score for stable interaction patterns and a short-window novelty score for sudden changes. LANTERN first filters unhelpful events, then selects a compact evidence set with Pareto-greedy selection, and finally adds one structure-aware analogical demonstration. Across ICEWS14, ICEWS05-15, ICEWS18, and GDELT, LANTERN consistently outperforms the state-of-the-art training-free baseline AnRe under the same backbone and 2-hop candidate protocol, improving Hits@1 by up to 2.5 points and MRR by up to 1.2 points.

问 Lune

几分钟读透这篇 Paper,不用再花一下午。

Lune 能读完全文,也能读引用它的顶会 Paper,讲清它的结论、证据和局限,每一点都注明出自哪篇 Paper。不用硬啃,也不靠猜。

可以从这些问题问起

Lune 可以查阅

全文 · 相关的 Paper

获取答案

免费开始,无需绑卡

脉络

  • Temporal Knowledge Graph Reasoning Based on Evolutional Representation Learning 引用了 Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks
  • Temporal Knowledge Graph Reasoning Based on Evolutional Representation Learning 引用了 Recurrent Event Network: Autoregressive Structure Inferenceover Temporal Knowledge Graphs
  • TimeTraveler: Reinforcement Learning for Temporal Knowledge Graph Forecasting 引用了 Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks
  • TimeTraveler: Reinforcement Learning for Temporal Knowledge Graph Forecasting 引用了 Recurrent Event Network: Autoregressive Structure Inferenceover Temporal Knowledge Graphs
  • TimeTraveler: Reinforcement Learning for Temporal Knowledge Graph Forecasting 引用了 Explainable Subgraph Reasoning for Forecasting on Temporal Knowledge Graphs
它引用的引用它的从左到右按年份排列,引用越多,位置越高。

它引用的 Paper16

  1. Recurrent Event Network: Autoregressive Structure Inferenceover Temporal Knowledge GraphsWoojeong Jin, Meng Qu, Xisen Jin, Xiang RenEMNLP 2020 · 353 次引用
  2. Temporal Knowledge Graph Reasoning Based on Evolutional Representation LearningZixuan Li, Xiaolong Jin, Wei Li, Saiping Guan 等SIGIR 2021 · 345 次引用
  3. Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation NetworksCunchao Zhu, Muhao Chen, Changjun Fan, Guangquan Cheng 等AAAI 2021 · 343 次引用
  4. Explainable Subgraph Reasoning for Forecasting on Temporal Knowledge GraphsZhen Han, Peng Chen, Yunpu Ma, Volker TrespICLR 2021 · 213 次引用
  5. TimeTraveler: Reinforcement Learning for Temporal Knowledge Graph ForecastingHaohai Sun, Jialun Zhong, Yunpu Ma, Zhen Han 等EMNLP 2021 · 164 次引用

相关 Paper

  1. AnRe: Analogical Replay for Temporal Knowledge Graph ForecastingGuo Tang, Zheng Chu, Wenxiang Zheng, Junjia Xiang 等ACL 2025 · 3 次引用
  2. From Implicit Graph Encoding to Explicit Evidence: A Training-Free LLM Framework for Temporal Knowledge Graph ReasoningGuo Tang, Ke Cheng, Huiming Fan, Heng Chang 等ACL Findings 2026
  3. Temporal Knowledge Graph Forecasting Without Knowledge Using In-Context LearningDong-Ho Lee, Kian Ahrabian, Woojeong Jin, Fred Morstatter 等EMNLP 2023 · 30 次引用
  4. Temporal Knowledge Graph Reasoning with Historical Contrastive LearningYi Xu, Junjie Ou, Hui Xu, Luoyi FuAAAI 2023 · 164 次引用
  5. Critic Rule Induction: Improving Temporal Knowledge Graph Forecasting with Generator-Critic Language ModelsYingsong Ning, Fu Zhang, Jingwei Cheng, Jiashun Peng 等ACL Findings 2026
黄昏的海面,两侧是细线勾勒的悬崖