Generating Streamlining Constraints with Large Language Models
This addresses the challenge of automating constraint generation for constraint programming practitioners, though it is incremental as it builds on existing LLM capabilities.
The paper tackles the problem of generating streamlining constraints for complex constraint satisfaction problems by using Large Language Models (LLMs) to propose effective streamliners, achieving substantial runtime reductions across seven diverse problems.
Streamlining constraints (or streamliners, for short) narrow the search space, enhancing the speed and feasibility of solving complex constraint satisfaction problems. Traditionally, streamliners were crafted manually or generated through systematically combined atomic constraints with high-effort offline testing. Our approach utilizes the creativity of Large Language Models (LLMs) to propose effective streamliners for problems specified in the MiniZinc constraint programming language and integrates feedback to the LLM with quick empirical tests for validation. Evaluated across seven diverse constraint satisfaction problems, our method achieves substantial runtime reductions. We compare the results to obfuscated and disguised variants of the problem to see whether the results depend on LLM memorization. We also analyze whether longer off-line runs improve the quality of streamliners and whether the LLM can propose good combinations of streamliners.