Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Video Text Generation on Video-ChatGPT benchmark v1 (test)

3.4Correctness of Information (CI)

GPT-4V

1.90242.29122.683.0688Mar 27, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.03
3.42.83.612.893.133.17
2024.03
3.262.763.572.343.283.04
2024.03
3.212.873.542.513.343.09
2024.03
3.172.793.522.513.253.05
2024.03
3.112.783.512.443.293.03
2024.03
3.022.883.512.662.812.98
2024.03
2.9633.532.462.512.89
2024.03
2.892.913.462.892.812.99
2024.03
2.762.933.012.242.422.67
2024.03
2.52.572.692.162.22.42
2024.03
2.442.643.182.262.312.57
2024.03
2.232.52.531.942.242.29
2024.03
2.032.322.31.982.152.16
2024.03
1.962.182.161.821.791.98