3 ms·
Fantastic article and impressive work by this company. They're basically wrapping LLM's with a working memory, and tying it to user input. And thus we step a li
by ypeterholmes 2y ago
Fantastic article and impressive work by this company. They're basically wrapping LLM's with a working memory, and tying it to user input. And thus we step a little closer to AGI/ASI.
- 0x008 2y agoIt’s probably Retrieval Augmented Generation? Just wanted to put this here in case somebody is not familiar with this pattern.