Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Coinflip

Benchmarks

Task NameDataset NameSOTA ResultTrend
Symbolic ReasoningCoinflip 4
Accuracy99.5
10
Judgment ConsistencyCoinFlip (unseen)
Baseline Score51.8
9
Symbolic ReasoningCoinFlip
Calibrated Accuracy100
8
Symbolic ReasoningCoinflip OOD: 4
Accuracy90.2
3
Symbolic ReasoningCoinflip OOD: 3
Accuracy98.6
3
Follow-up Questioning ConsistencyCoinFlip (unseen)
Baseline Consistency Score52.2
3
Symbolic ReasoningCoinflip OOD: 4 (test)
Accuracy-
0
Symbolic ReasoningCoinflip OOD: 3 (test)
Accuracy-
0
Showing 8 of 8 rows