Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces RALP, a method that reframes knowledge graph link prediction as a prompt learning problem for LLMs, enabling prediction of unseen entities, relations, and literals. RALP learns string-based chain-of-thought prompts as scoring functions for triples using Bayesian Optimization via the MIPRO algorithm, requiring only a few training examples. Experiments show RALP improves state-of-the-art KGE models by over 5% MRR and achieves over 88% Jaccard similarity on OWL reasoning tasks, demonstrating the effectiveness of prompt-based LLM reasoning.
LLMs can now reason over complex knowledge graphs with unseen entities and relations, outperforming traditional embedding methods, by learning to "think" through chain-of-thought prompts.
Knowledge graph embedding (KGE) models perform well on link prediction but struggle with unseen entities, relations, and especially literals, limiting their use in dynamic, heterogeneous graphs. In contrast, pretrained large language models (LLMs) generalize effectively through prompting. We reformulate link prediction as a prompt learning problem and introduce RALP, which learns string-based chain-of-thought (CoT) prompts as scoring functions for triples. Using Bayesian Optimization through MIPRO algorithm, RALP identifies effective prompts from fewer than 30 training examples without gradient access. At inference, RALP predicts missing entities, relations or whole triples and assigns confidence scores based on the learned prompt. We evaluate on transductive, numerical, and OWL instance retrieval benchmarks. RALP improves state-of-the-art KGE models by over 5% MRR across datasets and enhances generalization via high-quality inferred triples. On OWL reasoning tasks with complex class expressions (e.g., $\exists hasChild.Female$, $\geq 5 \; hasChild.Female$), it achieves over 88% Jaccard similarity. These results highlight prompt-based LLM reasoning as a flexible alternative to embedding-based methods. We release our implementation, training, and evaluation pipeline as open source: https://github.com/dice-group/RALP .