LLMs are great at understanding text This allows them to extract structured information from text: to use in forms, for query generation, knowledge base construction, etc `pip install kor` by @veryboldbagel is the best attempt at this I've seen eyurtsev.github.io/kor/index…

Mar 11, 2023 · 5:26 PM UTC

This is very neat. Where do you think this is best used? I can see some really good use cases for structuring LLM response data, which I found pretty annoying in vanilla JS. Allows you to get somewhat deterministic responses from LLMs without having to prompt engineer too much
I think this is slightly different than strucutring response data - agree that’s super useful and we’re working on some cool stuff there
Can this be used together with LangChain?
what would be the main way you'd want to use it? basically be able to treat this extraction pipeline as a chain?
This would be a beautiful way to structure the memory component of a conversational agent ⚡ Might not capture some subtleties but would probably work well with a summary memory or a buffer window memory to complement the parts that the knowledge base has missed.
Saved! Here's the compiled thread: mem.ai/p/fFmCbuAXu9czTFR86g6… 🪄 AI-generated summary: "Language models are powerful tools for understanding text, which can be used to extract structured information from text for various applications such as forms, query...
Saved! Here's the compiled thread: mem.ai/p/ebcy5uyzm3Zdu5CalkL… 🪄 AI-generated summary: "Language models are powerful tools for understanding text, which can be used to extract structured information from text for various applications such as forms, query...