original 1950496511550169518

Me hacen gracias estos "highlights", tipo GraphQL vs REST API. Creo que cada cosa tiene su caso. Igualmente, ¡gráficos chingones! Merece la pena echarle un ojo. https://t.co/OIBexFnsdU

Cita · 1950182433342841203 · Akshay 🚀 @akshay_pachaar · · en X
En respuesta a · 1950182409011671363 · Akshay 🚀 @akshay_pachaar · · en X

top-k retrieval in RAG rarely works.

Imagine you want to summarize a biography where each chapter details a specific accomplishment of an individual.

Traditional RAG struggles because it retrieves only top-k chunks while it needs the entire context. https://t.co/SBtO5Hx9lz

Graph RAG solves this by:

- Building a graph with entities and relationships from docs.
- Traversing the graph for context retrieval.
- Sending the entire context to the LLM for a response.

The visual shows how its different from naive RAG: https://t.co/JFYkot6xIv

Fuente verbatim: corpus/posts/1950496511550169518.md · en X · acto Julio 2025
Escrivivir · Scriptorium Skins · Animus Iocandi · Aleph Cero · F.A.R.O. · Material transmedia para agentes del juego ARG · AIGPL · Repositorio