Autocomplete

N-gram language models

The classic statistical baseline — predict the next token from the previous n−1.
RNNs (LSTM / GRU)

Recurrent models that carry a hidden state across the sequence.
CNNs for sequences

Convolutional models that read local windows in parallel.
Transformers

Self-attention over the whole sequence — the modern default.

No matches.

Press Esc or click anywhere to close