CRAFT Benchmark

Multi-agent coordination under partial information · 15 director models · 20 structures · 3 directors + 1 builder · oracle-assisted evaluation

arXiv 2603.25268 Dataset Code Nath, VanderHoeven, Krishnaswamy 2026
CRAFT benchmark overview
# Model Type Task progress ↓ IOU Oracle adh. Pos. acc. Complexity split