Benchmark RadarAI BENCHMARK PROFILE
PolyComp
PolyComp is a benchmark for compositional 3D spatial reasoning using polycube problems with 120 problems across four geometry families.
- Released
- 2026-08-13
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It stresses visual recognition and compositional reasoning, with random guessing baseline at 25% for challenging multimodal evaluation.
Motivation
We introduce PolyComp, a procedurally generated and verified benchmark that stresses visual recognition and compositional spatial reasoning.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.