gpt-score: 3.27
Papers with Code · CC BY-SA 4.0
Video-based Generative Performance Benchmarking (Contextual Understanding)
The benchmark evaluates a generative Video Conversational Model with respect to Contextual 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: 3.64
gpt-score: 3.46
gpt-score: 2.30
gpt-score: 3.57
gpt-score: 3.01
gpt-score: 3.9
gpt-score: 4.21
gpt-score: 3.84
gpt-score: 3.74
gpt-score: 3.86
gpt-score: 2.62
gpt-score: 2.16
gpt-score: 2.53
gpt-score: 3.74
gpt-score: 3.40