The FLIP benchmark of 11,674 blockchain image-story puzzles shows best open and closed AI models reach 75.5% and 77.9% accuracy, below the 95.3% human consensus baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FLIP Reasoning Challenge
The FLIP benchmark of 11,674 blockchain image-story puzzles shows best open and closed AI models reach 75.5% and 77.9% accuracy, below the 95.3% human consensus baseline.