A Solver-in-the-Loop Framework for Improving LLMs on Answer Set Programming for Logic Puzzle Solving

TL;DR

This paper introduces a solver-in-the-loop framework that enhances large language models' ability to generate Answer Set Programming code for logic puzzles by iterative fine-tuning guided by solver feedback.

Contribution

It presents a novel ASP-solver-in-the-loop approach for instruction-tuning LLMs, improving their semantic parsing capabilities for domain-specific code generation.

Findings

Consistent performance improvements on two datasets.

Effective use of solver feedback for data curation.

Enhanced robustness through solver-guided search.

Abstract

The rise of large language models (LLMs) has sparked interest in coding assistants. While general-purpose programming languages are well supported, generating code for domain-specific languages remains a challenging problem for LLMs. In this paper, we focus on the LLM-based generation of code for Answer Set Programming (ASP), a particularly effective approach for finding solutions to combinatorial search problems. The effectiveness of LLMs in ASP code generation is currently hindered by the limited number of examples seen during their initial pre-training phase. In this paper, we introduce a novel ASP-solver-in-the-loop approach for solver-guided instruction-tuning of LLMs to addressing the highly complex semantic parsing task inherent in ASP code generation. Our method only requires problem specifications in natural language and their solutions. Specifically, we sample ASP statements for program continuations from LLMs for unriddling logic puzzles. Leveraging the special property of declarative ASP programming that partial encodings increasingly narrow down the solution space, we categorize them into chosen and rejected instances based on solver feedback. We then apply supervised fine-tuning to train LLMs on the curated data and further improve robustness using a solver-guided search that includes best-of-N sampling. Our experiments demonstrate consistent improvements in two distinct prompting settings on two datasets.