Sequence Model
Also known as: Sequential Model, Sequence Model (RNN-style)
A sequence model processes ordered inputs one step at a time (or in a stack that respects order) and keeps a running summary of what it has seen so far—the hidden state.
Classic examples are RNNs, LSTMs, and GRUs reading tokens left to right. The same idea appears in speech and time series: at step $t$, the model combines the new observation $x_t$ with the previous summary $h_{t-1}$ to form $h_t$.
That running summary is powerful for short context and fragile for long context: early information can be overwritten. Soft attention and Transformers were invented largely to escape that pressure.