Streamlined Constraint Reasoning via CNN Pattern Recognition on Enumerated Solutions

TL;DR

This paper introduces a novel method combining CNN pattern recognition and LLM-driven synthesis to automatically generate effective streamliner constraints for constraint programming, significantly reducing search time.

Contribution

It presents a new approach that uses enumerated solutions and CNNs to inform LLM-based constraint generation, outperforming existing automated streamliner synthesis methods.

Findings

Achieved up to 98.8% reduction in portfolio time on benchmark models.

Discovered streamliners yield geometric-mean speedups of over 900x.

Effectively identifies class-based and canonicalization constraints.

Abstract

Constraint programming practitioners accelerate hard problems through a layered set of techniques applied in order of risk. Standard hardening (symmetry-breaking and implied constraints) is applied first and preserves satisfiability. Streamliner constraints, which restrict search to a structural sub-family of solutions, do not preserve satisfiability and are reserved as a final lever. Existing automated streamliner-synthesis approaches either search a constraint grammar or prompt a Large Language Model directly on the problem model. We propose a different approach: enumerate feasible solutions, train a Convolutional Neural Network contrastively against perturbed non-solutions to detect structural patterns, and translate the CNN's discriminative signal into candidate MiniZinc streamliners through LLM-driven synthesis. The CNN grounds the LLM's constraint generation in observed solution structure rather than model text alone. We evaluate on hardened benchmark models where streamliner discovery is the residual performance lever. Our pipeline achieves 98.8% portfolio time reduction on hardened Vessel Loading, 98.6% on hardened Social Golfers, and 89.4% on Black Hole, with best-single streamliners reaching geometric-mean speedups of 932x, 356x, and 1103x respectively. Discovered streamliners include class-based packing constraints on Vessel Loading, beyond-hardening canonicalisations on Social Golfers, and layout-coordinate bounds on Black Hole.