TreasureBench
Benchmark tasks
Every task declares what evidence is available, which actions are allowed, and how results are evaluated. The complete technical registry remains downloadable.
Scroll horizontally to see every column.
| Task | Family | Compatible strategy | Supporting studies | Claims |
|---|---|---|---|---|
| DB-G01 | canonical-atomic-boxes | blind-distinct, pooled-planner | DD-000 | DD-C-0003, DD-C-0006, DD-C-0008 |
| DB-G02 | private-team-role-counterexample | ex-ante-role-policy | DD-001 | DD-C-0021 |
| DB-G03 | canonical-private-team-optimum | private-clue-following | DD-001 | DD-C-0038 |
| DB-G04 | deterministic-disclosure-fixture | consensus | DD-002 | DD-C-0030 |
| DB-G05 | six-rule-selection-fixture | anonymous-market | DD-002 | DD-C-0040, DD-C-0041 |
| DB-G06 | homogeneous-source-network | private-clue-following | DD-003 | DD-C-0032, DD-C-0033 |
| DB-G07 | heterogeneous-source-moment-counterexample | posterior-sampling | DD-003 | DD-C-0043, DD-C-0044 |
| DB-G08 | perfect-elimination-sequential | sequential-greedy | DD-004 | DD-C-0045 |
| DB-G09 | overlapping-weighted-coverage | marginal-coverage-greedy | DD-005 | DD-C-0046 |
| DB-G10 | score-difference-mechanism | sole-rescue-response | DD-006 | DD-C-0048 |
| DB-G11 | dd006a-transfer-frontier | dd006a-mechanism | DD-006 | DD-C-0050 |
| DB-G12 | dd006b-strict-joint-mechanism | dd006b-mechanism | DD-006B | DD-C-0053 |
| DB-G13 | two-agent-evidence-acquisition | private-clue-following | DD-008 | DD-C-0051 |
| DB-G14 | n-agent-source-choice | ex-ante-role-policy | DD-008A | DD-C-0052 |
| DB-G15 | architecture-atlas | registered-atlas-architecture | DD-009 | DD-C-0054 |
| DB-G16 | dd012-attention-profiles | private-only, designated-reader, public-only | DD-012 | DD-C-0059 |
| DB-G17 | dd012-equilibrium-optimum-cell | voluntary-attention-equilibrium, designated-reader | DD-012 | DD-C-0060, DD-C-0061 |
| DB-G18 | dd013-audience-design | public-only, designated-reader, audience-optimal-assignment | DD-013 | DD-C-0062, DD-C-0063 |
| DB-G19 | dd013-implementation-mechanism | audience-optimal-assignment | DD-013 | DD-C-0064, DD-C-0065 |
| DB-G20 | dd014-conditional-policies | conditional-private-dominant, conditional-public-dominant, third-option-contrarian | DD-014 | DD-C-0066, DD-C-0067 |
| DB-G21 | dd016-threshold-discovery | threshold-tied-mode-selection, threshold-minimum-team-planner | DD-016 | DD-C-0071, DD-C-0073 |
| DB-G22 | dd017-threshold-equilibrium-registry | threshold-equilibrium-census | DD-017 | DD-C-0075, DD-C-0076, DD-C-0077, DD-C-0078 |
| DB-G23 | dd015-dynamic-attention | dynamic-autonomous-bayes, dynamic-common-information-planner | DD-015 | DD-C-0079, DD-C-0080, DD-C-0081 |
| DB-G24 | dd018-minimum-viable-team-mechanisms | team-token-mechanism, marginal-team-contribution, universal-pooling-team | DD-018 | DD-C-0083, DD-C-0084, DD-C-0085, DD-C-0086 |