Video Captioning on EVC datasets Human Evaluation 1.0 (test) benchmark leaderboard · SOTA2 Research

Video Captioning on EVC datasets Human Evaluation 1.0 (test)

7.22AccuracyVideoLLaMA3

Evaluation Results

Method	Links
VideoLLaMA32026.06		7.22	6.2	7.72	6.84
MM-ECPE++2026.06		7.02	6.98	8.44	7.76
GPT-4o2026.06		6.98	6.04	7.24	6.62
DCGN2026.06		6.88	6.43	7.68	6.32
MM-ECPE2026.06		6.4	6.75	8.04	7.36
EPAN2026.06		6.24	5.97	7.13	5.76
VideoBLIP2026.06		5.91	5.54	7.2	4.56
VEIN2026.06		5.47	5.18	6.27	5.04