Zubnet AILearnWiki › Encoder
Fundamentals

Encoder

Encoder Network, Feature Extractor
A neural network component that converts input data into a compressed, information-rich representation (encoding). In Transformers, the encoder uses bidirectional attention to process the full input and produce contextual representations. In autoencoders, the encoder compresses input into a latent bottleneck. In image generation, the VAE encoder converts images into latent space. Encoders are the "understanding" half of many architectures.

Why it matters

Encoders are everywhere: BERT is an encoder, CLIP has a text encoder and an image encoder, Stable Diffusion has a VAE encoder, RAG systems use encoder models for embeddings. Understanding what an encoder does — compresses input into a useful representation — helps you understand all of these systems. The quality of the encoding determines the quality of everything downstream.

Deep Dive

In a Transformer encoder (BERT, the left half of T5), every token attends to every other token bidirectionally. This means the representation of the word "bank" incorporates information from both "river" (left context) and "fishing" (right context) simultaneously. This bidirectional attention is why encoder representations are richer than decoder (left-to-right only) representations for understanding tasks.

Encoder vs. Decoder

The key distinction: encoders process input (understanding), decoders generate output (creation). Encoders see everything at once (bidirectional). Decoders see only past tokens (causal/left-to-right). This is why encoder models (BERT) are better for classification and search, while decoder models (GPT, Claude) are better for generation. Encoder-decoder models (T5, BART) use an encoder for input understanding and a decoder for output generation, connected by cross-attention.

Encoders in Multimodal Systems

Multimodal systems typically use separate encoders for each modality: a vision encoder (ViT) for images, a text encoder (BERT/CLIP) for text, and potentially audio encoders for speech. These produce embeddings in a shared space, enabling cross-modal understanding. The quality of each encoder determines how well the system understands that modality. This is why CLIP's training (aligning image and text encoders) was so impactful — it created a bridge between vision and language understanding.

Related Concepts

← All Terms
ESC