A way to shorten input text while keeping the key information.
Context Compression is like a friend recapping a three-hour movie in the popcorn line. You get the dragon fight, not the carpet color.
It shrinks long text before the AI reads it. You meet it in long document Q&A. It also helps Memory and huge chats.
Context-window
Context Compression uses shorter text to fit more information into the window.
RAG
RAG often uses it to shorten found sources before giving them to the model.
Memory
It turns long past memory into shorter information the AI can still use.
Token
Context Compression reduces the number of input tokens.