Base-to-New Generalization
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
51.58Base Performance
126
Jul 8, 2026
86.88Base Accuracy
114
Jul 2, 2026
82Base Accuracy
113
Jul 2, 2026
88.11Base Score
102
May 19, 2026
97.34Base Score
96
Jul 2, 2026
99.07Base Score
90
Jul 2, 2026
83.93Base Score
89
Jul 2, 2026
95.5Base Accuracy
71
Mar 19, 2026
57.84Base Accuracy
20
Apr 21, 2026
85.48Base Score
13
Feb 26, 2026
48.54Base Accuracy
10
Apr 21, 2026
52.17Base Accuracy
10
Apr 21, 2026
48Base Accuracy
10
Apr 21, 2026
32.84Base Set Score
10
Apr 21, 2026
68.94Base Accuracy
10
Apr 21, 2026
49.14Accuracy (Base)
10
Apr 21, 2026
76.07Accuracy (Base)
10
Apr 21, 2026
92.73Base Accuracy
10
Apr 21, 2026
98.57Base Accuracy
10
Feb 26, 2026
0.829Base Score
10
Feb 26, 2026
98.6Base Score
10
Feb 26, 2026
81.31Base Accuracy
10
Feb 26, 2026
57.87H Score
8
Jun 18, 2026
92.52Base Score
8
Feb 26, 2026
86.9Base Set Score
8
Feb 26, 2026