Thank you for share the knowledge!!!

I want ask for simple information:

Whether or not LSTM-Autoencoder is the "State of the art" method for reduce the dimension of sequence data? or there other promising research is alredy come out?

Thank you