Compositional Generalization
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
42.86Score
18
Feb 26, 2026
62.34Score
18
Feb 26, 2026
0.402Score
18
Feb 26, 2026
66.69Score
18
Feb 26, 2026
50Score
18
Feb 26, 2026
63.63Score
18
Feb 26, 2026
0.482Score
18
Feb 26, 2026
0.619Score
18
Feb 26, 2026
0.638Score
18
Feb 26, 2026
83.9Exact Match Accuracy
6
May 7, 2026
23.44Length
6
May 7, 2026
49P1 Score
6
Mar 11, 2026
100P1
6
Mar 11, 2026
84.7FID
5
May 11, 2026
62.6FID
5
May 11, 2026