About evaluations in your paper.

[osttkm]

Hi! thanks for your great work! I hava a question about experiments in your paper.

Regarding the experiments in Table 1 and Table 2, was the Otter model evaluated using the weights trained on the MIMIC-IT dataset? Or was it evaluated using weights that were additionally trained on a dataset like Imagenet? Please let me know.

[MacavityT]

Thanks for your attention, we used the weight from Oringin Weight, without any extra tunning.