Has anyone tried using an LLM for the retrieval stage? Instead of using vector e... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		thorum on Dec 25, 2023 \| parent \| context \| favorite \| on: Ask HN: How do I train a custom LLM/ChatGPT on my ... Has anyone tried using an LLM for the retrieval stage? Instead of using vector embeddings, have a (small, fast) LLM literally scan the entire corpus in chunks extracting relevant sections?

spullara on Dec 25, 2023 [–]

That would still be very slow with any reasonably large corpus.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact