This is the code repo for paper "VIBE: Video-to-Text Information Bottleneck Evaluation for TL;DR" - View it on GitHub