Video Captioning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
70.7CIDEr
16
Feb 26, 2026
74.1CIDEr
15
Feb 26, 2026
73CIDEr
15
Feb 26, 2026
96.6CIDEr
14
Feb 27, 2026
3.67Correctness
13
Feb 26, 2026
65.02CIDEr
12
Feb 27, 2026
65.13Entire Hallucination Score
11
Jul 7, 2026
54.8BLEU@4
11
Feb 27, 2026
43.64BLEU-1
11
Feb 26, 2026
14.48SODA_c
11
Feb 26, 2026
76.3CIDEr
11
Feb 26, 2026
71.5Audio Transcription Score
10
Jul 3, 2026
70.6Score
10
May 29, 2026
53.8Similarity Score
10
May 14, 2026
0.562Cost-H
10
Mar 27, 2026
2.35Score
10
Mar 27, 2026
7.67Relevance
10
Mar 27, 2026
7.38Relevance
10
Mar 27, 2026
72.67ICH-CC-en Score
10
Feb 26, 2026
27.9ROUGE-L
10
Feb 26, 2026
31.46ROUGE-L
10
Feb 26, 2026
54.9CIDEr
10
Feb 27, 2026
17.43BLEU@3
9
Feb 27, 2026
7.22Accuracy
8
Jun 9, 2026
47.8BLEU
7
Mar 4, 2026