ModelRefs / Late Interaction (ColBERT) — AI Glossary
Late Interaction (ColBERT) — AI Glossary
A retrieval architecture encoding query and document tokens separately, computing fine-grained token-level similarity at query time.
Overview
ColBERT (Khattab & Zaharia 2020) stores per-token embeddings for every document. At query time, each query token finds its best-matching document token (MaxSim). Late interaction achieves near cross-encoder quality with much faster search (ANN on token embeddings). ColBERT-v2 adds compression; RAGatouille provides a simple Python wrapper.
Reference details
| Topic | rag |
|---|---|
| Also known as | ColBERT, token-level interaction |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Late Interaction (ColBERT) — AI Glossary.
Frequently asked questions
What is Late Interaction (ColBERT)?
A retrieval architecture encoding query and document tokens separately, computing fine-grained token-level similarity at query time.
Is Late Interaction (ColBERT) the same as ColBERT?
Yes — ColBERT, token-level interaction are common aliases for Late Interaction (ColBERT).
What concepts are related to Late Interaction (ColBERT)?
Closely related concepts include bi encoder, dense retrieval, reranking.