RAG, Fine-Tuning & ContextOmar Khattab et al. (Stanford)Stanford University

Khattab et al. present ColBERT

Presented ColBERT — keeps queries and documents as token-level embeddings, then scores them with MaxSim (late interaction)

Why it matters

The starting point of the late interaction lineage, distinct from single-vector retrieval

Source — Primary sources first; secondary reports are labelled “Press”

Paperarxiv.orghttps://arxiv.org/abs/2004.12832

Added in Week 35, 2026 · See that week

Nearby on the same axis

FAIR, UCL and NYU publish the original RAG paperPaperLoRAPaperSanthanam et al. present ColBERTv2PaperChase publishes the first public release of LangChain (PyPI 0.0.1)Official