The information I can see from the article is that there is no essential difference between the proposed Selective Token Mixer and mamba. Regarding Selective Channel Mixer, I understand that it operates on the channel through fliping, but it is not clearly stated in the article. In addition, as can be seen from Figure 1 in the article, the designs of Selective Token Mixer and Selective Channel Mixer are the same. Are these two designs the same? I hope the author will release the code as soon as possible to resolve doubts.
The information I can see from the article is that there is no essential difference between the proposed Selective Token Mixer and mamba. Regarding Selective Channel Mixer, I understand that it operates on the channel through fliping, but it is not clearly stated in the article. In addition, as can be seen from Figure 1 in the article, the designs of Selective Token Mixer and Selective Channel Mixer are the same. Are these two designs the same? I hope the author will release the code as soon as possible to resolve doubts.