// In-context Retrieval at Million-token Scale // Great study providing better understanding of retrieval at million-token scale. They run t…

// In-context Retrieval at Million-token Scale // Great study providing better understanding of retrieval at million-token scale. They run the first systematic study of in-context retrieval at the scales real retrievers face, million-token corpora and length generalization far beyond training size. They introduce BlockSearch, a 0.6B LM retriever whose architectural and training changes improve over prior LM baselines and length-generalize up to 10 times beyond its training length. Paper: https://arxiv.org/abs/2607.01538 Learn to build effective AI agents in our academy: https://academy.dair.ai/

Source: DAIR.AI (X) | 2026-07-06