The difference to RNNs is that the state in SSMs is a linear combination of previous inputs, which makes it possible to parallelise training in various ways. G

The difference to RNNs is that the state in SSMs is a linear combination of previous inputs, which makes it possible to parallelise training in various ways. Gates and non-linearities in RNNs allow the state to be "any" function of previous inputs.