Search papers, labs, and topics across Lattice.
This paper introduces ADVENT, an LLM-driven mechanism for automatic predicate invention (PI) in Inductive Logic Programming (ILP), addressing the limitations of existing methods that rely on domain expertise and produce semantically opaque predicates. By combining LLM abductive generation with Prolog deductive verification in an iterative loop, ADVENT refines candidate predicates based on concrete execution results, leading to the invention of meaningful and interpretable predicates. Experimental results demonstrate that ADVENT achieves a 58% success rate in predicate invention where traditional ILP fails, improving to 80% with formal verification and yielding significant gains in cross-task knowledge reuse.
LLM-driven predicate invention can achieve an 80% success rate in ILP, transforming a traditionally expert-dependent process into an automated one.
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.
