Yingshu Li
Yingshu Li
Have you solved the problem? I have the same problem with you as the BLEU-4 score capped at about 15-16.
>  Professor, is there a problem with my classifier loss？ The same as your loss. Is this one correct?
> Well, is this with the noisy images? This is possible, but just check the performance during image-to-image translation Thanks for your reply. Another question is I found you print...
> Hi! I can take this. > > I reproduced the discrepancy with the provided snippet. The gap stems from tokenization differences: evaluate’s BLEU uses SacreBLEU-style tokenization (e.g., 13a), while...
In your r2gencsr: We utilize the Findings as our ground truth and randomly partition the dataset into a ratio of 7:1:2, which consists of training, testing, and validation sets with...
> Our division principle follows the approach of R2GenCSR. The result of the division is that the training set: validation set: test set =40463:5780:11562. This is a clerical error in...