RAG, Fine-Tuning & ContextShi-Qi Yan et al. (University of Science and Technology of China, UCLA)

Corrective RAG published

Corrective RAG published — an evaluator judges retrieval results as correct/incorrect/ambiguous and corrects them with web search

Why it matters

A defensive RAG design that assumes retrieval will fail

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

Paperarxiv.orghttps://arxiv.org/abs/2401.15884

Added in Week 35, 2026 · See that week

Same week

KTO — alignment from binary signals based on prospect theoryPaperNomic Embed publishedPaperRAPTOR publishedPaperDolmaPaper

Nearby on the same axis

RAPTOR publishedPaperGemini 1.5 ProOfficialAnthropic releases Claude 2.1 with a 200K-token contextOfficialGreg Kamradt publishes the needle-in-a-haystack long-context recall stress testPress