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
TemporalSelfAttention does not address the issue of extracting indices from value based on query. There is still flaw here.
During training mode, in the TemporalSelfAttention module, the generation of the first frame's query takes the previous frame's value. However, this value is the original query without encoding. Therefore, the historical queries of the first frame actually lack positional encoding. Shouldn't they be added?
The text was updated successfully, but these errors were encountered:
TemporalSelfAttention does not address the issue of extracting indices from value based on query. There is still flaw here.
During training mode, in the TemporalSelfAttention module, the generation of the first frame's query takes the previous frame's value. However, this value is the original query without encoding. Therefore, the historical queries of the first frame actually lack positional encoding. Shouldn't they be added?
The text was updated successfully, but these errors were encountered: