gpt-score: 2.34
Papers with Code · CC BY-SA 4.0
Video-based Generative Performance Benchmarking (Temporal Understanding)
The benchmark evaluates a generative Video Conversational Model with respect to Temporal Understanding.
We curate a test set based on the ActivityNet-200 dataset, featuring videos with rich, dense descriptive captions and associated question-answer pairs from human annotations. We develop an evaluation pipeline using the GPT-3.5 model that assigns a relative score to the generated predictions on a scale of 1-5.
논문
gpt-score: 2.66
gpt-score: 2.39
gpt-score: 1.98
gpt-score: 2.65
gpt-score: 2.24
gpt-score: 2.67
gpt-score: 3.21
gpt-score: 2.77
gpt-score: 2.93
gpt-score: 2.77
gpt-score: 1.98
gpt-score: 1.82
gpt-score: 1.94
gpt-score: 2.83
gpt-score: 2.49