Transformers and LLMs

Short Convolution, explained

A short convolution is a small local convolution used to mix nearby sequence information efficiently.
Back to Glossary

Part of the Semantic Notion technical glossary.

Short Definition

A short convolution is a small local convolution used to mix nearby sequence information efficiently.

Intuition

Some local structure does not need full attention. A short convolution can cheaply blend neighboring hidden states.

Technical Definition

Short convolution applies a learned one-dimensional kernel over a bounded sequence window, usually over hidden states or token representations.

Example

Inkling uses SConv over hidden states alongside attention layers for local mixing.

Common Misunderstandings

Short convolution is not a replacement for all attention.
Its benefit depends on how it is integrated with the rest of the architecture.

Same Category

More from Transformers and LLMs

AttentionAttention is a neural network mechanism that lets a model decide which parts of an input are most relevant when producing a representation or output.Context WindowA context window is the maximum amount of input and generated tokens a model can consider at one time.Cross-AttentionCross-attention lets one sequence attend to information from a different sequence or modality.Hybrid AttentionHybrid attention combines different attention patterns, such as local and global attention, inside one model.KV CacheA KV cache stores attention keys and values from previous tokens so autoregressive generation does not recompute them every step.Multimodal ModelA multimodal model can process or generate more than one kind of data, such as text, images, audio, or video.

Related Reading

Essays that use this concept

Transformers and LLMsInkling: What Thinking Machines Actually Shipped

Start here

Need the broader concept map?

Return to the glossary index to move from this definition into adjacent AI, ML, mathematics, and computer science terms.

Browse Glossary