Counterfactual Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
2.15Diversity
5
May 29, 2026
1.44Diversity
5
May 29, 2026
2.2Diversity
5
May 29, 2026
0.09MAE (b)
5
May 8, 2026
99.26F1 Score (Generated)
5
May 8, 2026
8.3FID
5
Feb 26, 2026
0.36L1 Loss
5
Feb 26, 2026
0.42L1 Distance
5
Feb 26, 2026
0.79L1 Distance
5
Feb 26, 2026
94.7Success Rate
4
Jul 3, 2026
1.36Diversity
4
May 29, 2026
0.184Composition MAE
4
Mar 25, 2026
0.058MAE (distance)
4
Feb 26, 2026
0.66Validity
3
Jun 11, 2026
85Retained Percentage
3
Jun 2, 2026
99.5Retention Rate
3
Jun 2, 2026
0.68Runtime (mins)
3
Jun 2, 2026
0.91Latency (minutes)
3
Jun 2, 2026
100Retained (%)
3
Jun 2, 2026
97.8Retention Rate
3
Jun 2, 2026
99.9Retained
3
Jun 2, 2026
45VCR Score
3
Jun 2, 2026
90Diversity
3
May 29, 2026
84Diversity
2
May 29, 2026
0.024MAE (d)
2
Feb 26, 2026