Why might reimplementation results in higher F1 scores than the reported baseline?

Hello everyone, While reimplementing a video summarization model, I noticed something unexpected: my reproduced results give higher F1 scores than the baseline reported in the original paper. I...

last month 2,503 2 View