CoinRAG: Contextualized Information Nugget KV Cache Reuse for Long-Context RAG

Name: CoinRAG: Contextualized Information Nugget KV Cache Reuse for Long-Context RAG
Uploaded: 2026-08-11T04:05:49.799644+00:00
Duration: 1181 s
Description: This episode discusses CoinRAG, a method for faster retrieval-augmented generation. It precomputes key-value caches for text chunks offline, then at query time selects tiny evidence spans ('nuggets') and slices their cached representations, avoiding re-encoding. Under a 100ms latency budget, CoinRAG outperforms chunk-level baselines like TurboRAG, achieving higher F1 with shorter contexts.

This episode discusses CoinRAG, a method for faster retrieval-augmented generation. It precomputes key-value caches for text chunks offline, then at query time selects tiny evidence spans ('nuggets') and slices their cached representations, avoiding re-encoding. Under a 100ms latency budget, CoinRAG outperforms chunk-level baselines like TurboRAG, achieving higher F1 with shorter contexts.

Open the full episode