(blog.voyageai.com)

261 points fzliu | 2 comments | 17 Nov 24 07:42 UTC | HN request time: 0.48s | source

1. djoldman ◴[17 Nov 24 13:09 UTC] No.42164014[source]▶

This is a cool way to look at multimodal embeddings. They look at performance as the the percentage of inputs slides from one modality to another:

replies(1): >>42189246 #

2. swyx ◴[19 Nov 24 23:37 UTC] No.42189246[source]▶

why does it pop up at the end?

All-in-one embedding model for interleaved text, images, and screenshots