Proceedings of the 2024 XCSP3 Competition
Gilles Audemard, Christophe Lecoutre, Emmanuel Lonca
TL;DR
The XCSP$^3$ Competition 2024 proceedings document a problem-selection framework emphasizing novelty, constraint diversity, and solver-scaling evaluation, with off-competition notes for ACE due to difficulty-bias concerns. The main contribution is a largely new set of CSP/COP problems implemented primarily in PyCSP$^3$ with reproducible artifact archives and a 34-problem main-track catalog drawn from the XCSP$^3$-core constraints. The paper also details the AverageAvoiding problem family, including its description, instance data across a range of $n$, and a concrete PyCSP$^3$ model that uses AllDifferent, a symmetry-break, and specific arithmetic constraints to enforce the problem property. Overall, the work provides a structured, reproducible benchmark suite and methodology for evaluating constraint solvers on diverse, scalable problems, aiding solver comparison and future competition design.
Abstract
This document represents the proceedings of the 2024 XCSP3 Competition. The results of this competition of constraint solvers were presented at CP'24 (30th International Conference on Principles and Practice of Constraint Programming).