ADVENT: LLM-Driven Automatic Predicate Invention for ILP

2026-07-01

AI Summary

This work addresses the limitations of traditional inductive logic programming (ILP), where predicate invention is typically expert-dependent, semantically opaque, and difficult to reuse across tasks. The paper introduces a novel framework that integrates large language models (LLMs) into predicate invention for the first time, establishing an iterative closed-loop process that combines LLM-based abductive generation with Prolog-based deductive verification. Guided by execution feedback, the LLM automatically refines semantically clear and interpretable new predicates, which are then stored in a reusable knowledge pool. Evaluated on nine poker hand concept learning tasks, the approach achieves an 80% success rate—significantly outperforming pure ILP baselines—and demonstrates cross-task reuse gains of up to 31 percentage points.

Abstract

Predicate invention (PI), the creation of new predicates to extend the hypothesis space, remains a critical bottleneck in Inductive Logic Programming (ILP). Existing methods rely on domain expertise and produce semantically opaque predicates, hindering adaptation to unfamiliar domains and cross-task reuse. We present ADVENT, an LLM-driven PI mechanism for ILP. ADVENT pairs LLM abductive generation with Prolog deductive verification, forming an iterative loop in which concrete execution results guide the LLM to refine candidate predicates. The mechanism leverages Large Language Models to identify implicit patterns in structured relational data and invent auxiliary predicates with meaningful names and definitions. Invented predicates and learned rules accumulate in a knowledge pool for cross-task reuse. Experiments on nine poker-hand concepts across seven LLMs show that LLM-driven PI achieves 58% success rate where ILP alone fails entirely, formal verification raises this to 80%, and the knowledge pool yields gains up to +31 percentage points, while producing human-interpretable rules. These results suggest that ADVENT offers a promising direction for automating predicate invention and enabling cross-task knowledge reuse in ILP.

Problem

• Predicate Invention
• Inductive Logic Programming
• Knowledge Reuse
• Semantic Opacity
• Hypothesis Space

Innovation

• Predicate Invention
• Inductive Logic Programming
• Large Language Models
• Abductive Reasoning
• Knowledge Reuse