Factual Correction
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
97ROUGE
25
Feb 26, 2026
58.2CR
14
May 8, 2026
74.79Factual Correction Score (GPT-4V)
6
Feb 26, 2026
52.35GPT-4V Score
6
Feb 26, 2026
61.34Factual Correction Score (GPT-4V)
6
Feb 26, 2026