Streamliners for Answer Set Programming

TL;DR

This paper adapts the StreamLLM approach to Answer Set Programming, using Large Language Models to generate streamliners that significantly speed up solving combinatorial problems.

Contribution

It introduces a method for automatically generating effective streamliner constraints for ASP using LLMs, improving solving efficiency on benchmark problems.

Findings

Speedups of up to 4-5x over original encodings on benchmarks.

Different LLMs produce diverse, semantically meaningful constraints.

The approach captures genuine problem structure beyond syntactic variations.

Abstract

Streamliner constraints reduce the search space of combinatorial problems by ruling out portions of the solution space. We adapt the StreamLLM approach, which uses Large Language Models (LLMs) to generate streamliners for Constraint Programming, to Answer Set Programming (ASP). Given an ASP encoding and a few small training instances, we prompt multiple LLMs to propose candidate constraints. Candidates that cause syntax errors, render satisfiable instances unsatisfiable, or degrade performance on all training instances are discarded. The surviving streamliners are evaluated together with the original encoding, and we report results for a virtual best encoding (VBE) that, for each instance, selects the fastest among the original encoding and its streamlined variants. On three ASP Competition benchmarks (Partner Units Problem, Sokoban, Towers of Hanoi), the VBE achieves speedups of up to 4--5x over the original encoding. Different LLMs produce semantically diverse constraints, not mere syntactic variations, indicating that the approach captures genuine problem structure.