cs.LGNov 7, 2026

HexIt: Learning Hexahedral Block Decomposition with Expert Iteration on Combinatorial Maps

Authors: Arjun Narayanan, Per-Olof Persson

Organizations: University of California, Berkeley · Department of Mathematics, University of California, Berkeley

Abstract

Decomposing a polyhedral solid into a few well-shaped hexahedral blocks is a long-standing open problem in mesh generation. We cast it as a sequential decision problem on a 3-dimensional generalised map: an agent splits edges, faces and volumes with three insert operations until every cell is a hexahedron and the mesh passes a quality threshold, and a policy that convolves over the map's involutions chooses each insert. A successful decomposition takes over a hundred decisions, and a uniformly random policy never finds one. We clone a policy from constructed decompositions and improve it by expert iteration. On held-out chamfered solids that share no boundary topology with the cloning data, the final policy solves an expected 94.4%94.4\% in 2020 samples, and expert iteration raises the best-of-20 solve rate on large chamfered solids from 0.2030.203 to 0.3550.355. At matched compute, PPO, GRPO and VinePPO updates all make the cloned policy worse.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Topology-Aware Reinforcement Learning for CAD Decomposition toward All-Hexahedral Meshing

    Nov 5, 2026Jiaming Peng, Zhihao Zheng, Rui Wang +53D Mesh GenerationReinforcement Learning

  2. MeshOctave: Vertex Split-and-Rewire Cascades for Native Mesh Generation

    Sep 30, 2026Junkai Lin, Tianhao Zhao, Hang Long +103D Mesh GenerationAdaptive Mesh Refinement