Open ZijinQiu opened 2 weeks ago
Please refer to this issue
Thank you for your reply. When reproducing the code, we found that there are data such as near_x_min, etc. These four data are not explained in the paper. What do they mean? I hope to get your help~
And what model is used to generate the intelligent annotations in the article? T5?
Noticed in the paper that there are only 4k manually labelled data, aren't there more than 5k, easily available, thanks!