gpt-score: 2.46
Papers with Code · CC BY-SA 4.0
Video-based Generative Performance Benchmarking (Consistency)
The benchmark evaluates a generative Video Conversational Model with respect to Consistency.
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: 2.81
gpt-score: 2.81
gpt-score: 2.15
gpt-score: 2.67
gpt-score: 2.42
gpt-score: 3.25
gpt-score: 3.81
gpt-score: 3.57
gpt-score: 2.81
gpt-score: 3.69
gpt-score: 2.37
gpt-score: 1.79
gpt-score: 2.24
gpt-score: 3.39
gpt-score: 2.47