This paper presents an exactly solvable discovery benchmark with sixteen boxes, one target, eight searchers, and noisy private clues. Pooling information raises the accuracy of the best single recommendation from 0.20 to 0.3835, yet a one-answer protocol causes everyone to repeat it, reducing group discovery from 0.8322 to 0.3835. An eight-action coordinated portfolio reaches 0.8594, while seven coordinated actions recover the decentralized benchmark. The paper also analyzes prize-splitting incentives, correlated reports, and proportional large-market limits.
No heat snapshots are available in the last 24 hours.