Captioning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
67.69Wtd Score
23
Jun 4, 2026
83.5Single Score
23
Jun 1, 2026
61.5METEOR
19
Jun 23, 2026
13.5Accuracy
16
Jun 18, 2026
50.8Accuracy
16
Jun 18, 2026
68.11Wtd
16
Jun 4, 2026
77.07Wtd
16
Jun 4, 2026
0.253AURC (CIDEr-N)
15
Feb 26, 2026
0.158AURC (CIDEr-N)
15
Feb 26, 2026
10.08BLEU-4
13
Apr 10, 2026
25.05AR-Score
11
Apr 20, 2026
53.32BLEU-1
10
Feb 26, 2026
26.54AR-Score
9
Apr 20, 2026
46.2BLEU-1
9
Mar 25, 2026
38BLEU
9
Mar 16, 2026
44BLEU
9
Mar 16, 2026
44BLEU
9
Mar 16, 2026
62.8CIDEr
9
Feb 26, 2026
60.3B4 Score
8
Jun 24, 2026
47.25Wtd
8
Jun 4, 2026
48.74Wtd Score
8
Jun 4, 2026
50.25Wtd Score
8
Jun 4, 2026
48.67Wtd Score
8
Jun 4, 2026
0.44ROUGE-L
8
May 22, 2026
4.03CI
8
May 20, 2026