Sign in Subscribe

Plain English Papers

Snapchat used AI agents to build a sound-aware video captioning system

They used "teachers" to describe 3.8M videos for their Panda-70M dataset

Snapchat used AI teachers to caption millions of videos clips

Read next

Encoder architecture from the research paper.

How Sora (actually) works

A change to make diffusion models 80% faster

A change to make diffusion models 80% faster

New discovery: leaner, meaner diffusion

Think differently!

The death of creativity