gpt-score: 2.68
Papers with Code · CC BY-SA 4.0
Video-based Generative Performance Benchmarking (Correctness of Information)
The benchmark evaluates a generative Video Conversational Model with respect to Correctness of Information.
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: 3.40
gpt-score: 2.89
gpt-score: 2.03
gpt-score: 3.08
gpt-score: 2.76
gpt-score: 3.60
gpt-score: 3.85
gpt-score: 3.48
gpt-score: 3.23
gpt-score: 3.55
gpt-score: 2.40
gpt-score: 1.96
gpt-score: 2.32
gpt-score: 3.27
gpt-score: 2.78