Streamliners for Answer Set Programming

2026-04-21

AI Summary

This work addresses the inefficiency of solving combinatorial problems in Answer Set Programming (ASP) by introducing, for the first time, a large language model (LLM)-based streamlining approach adapted from constraint programming. By designing targeted prompts, the method guides the LLM to automatically generate semantically diverse and non-redundant structured constraints. These candidate constraints are dynamically filtered through ASP encoding analysis and validity verification, enabling the construction of virtually optimal encoding variants across multiple benchmarks. Evaluated on three ASP competition benchmarks, the approach achieves up to a 4–5× speedup in solving time, significantly enhancing computational efficiency and demonstrating the effectiveness and generalizability of LLM-driven streamlining in ASP.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Programming
• Search and Optimization: Combinatorial Optimization
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

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.

Problem

• Answer Set Programming
• Streamliner constraints
• Combinatorial problems
• Search space reduction
• Large Language Models

Innovation

• Answer Set Programming
• Streamliner constraints
• Large Language Models
• Combinatorial problem solving
• Search space reduction