Efficient Multi-Vector Retrieval with Muvera [Podcast]

2 points

a year ago

Multi-Vector Retrieval methods, such as ColBERT and ColPali, offer powerful search capabilities by combining the power of cross encoders and learned representations. This is achieved with the late interaction distance function and contextualized token embeddings. However! The associated costs of storing, indexing, and searching these expanded vector representations is a major challenge!

Enter MUVERA! MUVERA introduces a novel compression technique specifically designed to make multi-vector retrieval more efficient and scalable!

This podcast begins with a primer on Multi-Vector Retrieval methods and then dives deep into the inner workings of MUVERA! I hope you find it useful, as always more than happy to discuss these ideas further with you!

YouTube: https://www.youtube.com/watch?v=nSW5g1H4zoU

Spotify: https://creators.spotify.com/pod/show/weaviate/episodes/MUVERA-with-Rajesh-Jayaram-and-Roberto-Esposito---Weaviate-Podcast-123-e33fnpi