WEBVTT

00:01:30 --> 00:03:06
Overcoming large language model knowledge gaps

00:03:06 --> 00:04:26
Understanding the architecture of retrieval augmented generation

00:04:26 --> 00:05:28
Searching by semantic meaning with vector embeddings

00:05:28 --> 00:11:42
Creating a sparse bag-of-words vector from plain text

00:11:42 --> 00:17:39
Measuring dimensional vector angle with cosine similarity

00:17:39 --> 00:19:38
Returning relevant matches with local similarity sorting

00:19:38 --> 00:21:16
Understanding the performance and semantic limits of bag-of-words

00:21:16 --> 00:22:53
Switching from sparse representations to dense embedding models

00:22:53 --> 00:26:04
Scaling vector similarity search using dedicated databases

00:26:04 --> 99:59:59
Advancing semantic boundaries with colbert and knowledge graphs