Comparison
Decoder vs Encoder-Decoder
Decoder and Encoder-Decoder are both common AI/LLM terms but cover different ideas. Here is a quick side-by-side.
When you would reach for Decoder
Decoder comes up when the question is fundamentally about architecture.
GPT-4 generating a paragraph token by token.
When you would reach for Encoder-Decoder
Encoder-Decoder comes up when the question is fundamentally about architecture.
T5: every NLP task framed as text-to-text.
Frequently asked
What is the difference between Decoder and Encoder-Decoder?
Decoder: A decoder is a transformer module that generates a sequence one token at a time, using causal self-attention so each token only sees earlier ones. GPT-style LLMs are decoder-only. Encoder-Decoder: An encoder-decoder model has a separate encoder that reads the input and a decoder that generates the output, with cross-attention linking them. T5 and the original transformer are encoder-decoders.
When should I use Decoder vs Encoder-Decoder?
Decoder is the right concept when you are focused on architecture. Encoder-Decoder applies when you are focused on architecture.
Are Decoder and Encoder-Decoder the same thing?
No. Decoder is architecture; Encoder-Decoder is architecture. They are related but address different parts of the AI stack.