RECIPE-TKG: From Sparse History to Structured Reasoning for LLM-based Temporal Knowledge Graph Completion

TL;DR

RECIPE-TKG is a novel, data-efficient framework that enhances temporal knowledge graph completion by combining rule-based retrieval, contrastive fine-tuning, and semantic filtering, outperforming previous LLM-based methods especially with sparse historical data.

Contribution

It introduces a lightweight, structured approach integrating retrieval, contrastive learning, and filtering to improve TKG completion with limited historical evidence.

Findings

Achieves up to 30.6% relative improvement in Hits@10

Produces more semantically coherent predictions

Effective in scenarios with sparse historical context

Abstract

Temporal Knowledge Graphs (TKGs) represent dynamic facts as timestamped relations between entities. TKG completion involves forecasting missing or future links, requiring models to reason over time-evolving structure. While LLMs show promise for this task, existing approaches often overemphasize supervised fine-tuning and struggle particularly when historical evidence is limited or missing. We introduce RECIPE-TKG, a lightweight and data-efficient framework designed to improve accuracy and generalization in settings with sparse historical context. It combines (1) rule-based multi-hop retrieval for structurally diverse history, (2) contrastive fine-tuning of lightweight adapters to encode relational semantics, and (3) test-time semantic filtering to iteratively refine generations based on embedding similarity. Experiments on four TKG benchmarks show that RECIPE-TKG outperforms previous LLM-based approaches, achieving up to 30.6\% relative improvement in Hits@10. Moreover, our proposed framework produces more semantically coherent predictions, even for the samples with limited historical context.