RAG, Fine-Tuning & ContextParth Sarthi et al. (Stanford)Stanford University

RAPTOR published

RAPTOR published — recursively summarizes documents into a tree and retrieves at multiple levels of abstraction

Why it matters

Proposed a hierarchical summary index as an alternative to chunk-level retrieval

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

Paperarxiv.orghttps://arxiv.org/abs/2401.18059

Added in Week 35, 2026 · See that week

Same week

KTO — alignment from binary signals based on prospect theoryPaperNomic Embed publishedPaperDolmaPaperCorrective RAG publishedPaper

Nearby on the same axis

Corrective RAG publishedPaperGemini 1.5 ProOfficialRULER publishedPaperAnthropic releases Claude 2.1 with a 200K-token contextOfficial