Predicate Renaming via Large Language Models

2025-10-29

AI Summary

This work addresses the unnamed predicate problem in Inductive Logic Programming (ILP) arising from predicate invention, which severely impairs the readability, interpretability, and reusability of learned logical rules. We propose the first large language model (LLM)-based semantic predicate naming method, leveraging LLMs’ joint capabilities in natural language understanding and logical code comprehension to semantically parse the definitional context of unnamed predicates and generate semantically accurate, domain-convention-compliant candidate names. Evaluated on a manually curated logic rule dataset, our method significantly outperforms baseline approaches, consistently producing highly relevant and human-understandable predicate names. This study pioneers the application of LLMs to the logical predicate naming task, establishing a novel paradigm for explainable AI and human-AI collaborative construction of logical theories.

Technology Category

• Machine Learning: Statistical Relational/Logic Learning
• Knowledge Representation and Reasoning: Logic Programming
• Natural Language Processing: (Large) Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

In this paper, we address the problem of giving names to predicates in logic rules using Large Language Models (LLMs). In the context of Inductive Logic Programming, various rule generation methods produce rules containing unnamed predicates, with Predicate Invention being a key example. This hinders the readability, interpretability, and reusability of the logic theory. Leveraging recent advancements in LLMs development, we explore their ability to process natural language and code to provide semantically meaningful suggestions for giving a name to unnamed predicates. The evaluation of our approach on some hand-crafted logic rules indicates that LLMs hold potential for this task.

Problem

• Renaming unnamed predicates in logic rules
• Improving readability and interpretability of logic theories
• Using LLMs for semantic predicate naming suggestions

Innovation

• Renaming predicates using large language models
• Generating meaningful names for logic rules
• Applying LLMs to improve logic theory readability