ColBERT / Late Interaction / Rerankers / kNN-LM

Editorial record

Plain-language summary

ColBERT represents each query and document as a bag of contextual token embeddings and computes relevance by summing, for each query token, its maximum similarity to any document token, so most document computation is precomputed offline and only lightweight matching happens at query time. This late-interaction design, alongside related rerankers and nearest-neighbor language models, delivered near cross-encoder ranking quality at a scale and latency close to bag-of-vectors search, making high-quality dense retrieval practical over large collections.