Small-to-big retrieval is underrated: embed the precise bit, expand after a hit
yi_ding · x · 2026-10-09
In a retweeted tip, yiding highlights an underrated RAG practice: small-to-big retrieval — embed the precise chunk, then expand to the larger context only after a hit. Embedding small improves retrieval precision without sacrificing the context completeness fed to the model.
More from coding & agent
- When sub-agents are actually worth it: fan-out research and pre-planned parallel tasks — brandon_galang · 2026-10-09
- 2026 tech industry report: nobody hand-writes code, engineers run 5-10 agents — bibryam · 2026-10-09
- Pydantic AI ships On-Demand Capabilities: instructions, tools and hooks triggered just in time — samuelcolvin · 2026-10-09
- Simple fix for agents treating other agents' decisions as sacrosanct: demand verbatim quotes — menhguin · 2026-10-09
- HQ hits 1,000 companies, ships desktop app and v16 with background workers on Claude, Codex, Grok — jacob_posel · 2026-10-09
- Open-source agent supervisor Foreman adds LangChain Deep Agents support — hwchase17 · 2026-10-09