RAG, Fine-Tuning & ContextBrian J Chan et al. (National Chengchi University, Academia Sinica)
Brian J Chan et al. present Cache-Augmented Generation
Cache-Augmented Generation presented — knowledge is placed in the context in advance and the KV cache is preloaded, eliminating the retrieval step
Why it matters
A branch that skips RAG altogether in the long-context era
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2412.15605Added in Week 35, 2026 · See that week