You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi! Thank you for your great work!
I have a question that hope you can explain. As your paper shows that the SS2D includes 3 phases: cross-scan, S6 blocks, and cross-merge. However, I can not find the S6 block explicitly in SS2Dv2 code. So is the SSM implemented inside the selective_scan function? (which I cannot access to).
The text was updated successfully, but these errors were encountered:
Yes. The S6 block is mainly adopted from tridao's mamba, so does the cuda implementation. And we have only some slight modifications on it. You can find the detailed implementation in kernels/selective_scan.
Hi! Thank you for your great work!
I have a question that hope you can explain. As your paper shows that the SS2D includes 3 phases: cross-scan, S6 blocks, and cross-merge. However, I can not find the S6 block explicitly in SS2Dv2 code. So is the SSM implemented inside the selective_scan function? (which I cannot access to).
The text was updated successfully, but these errors were encountered: