AI Roundup 06/08-> 06/15/2023

Meta turns text and melodies into new music, Microsoft's vision-language model that understands the world, and a novel dataset for autonomous agents to navigate the web

1. Meta AI releases MusicGen - a simple and controllable model for music generation using text prompts and input melodies (TechCrunch)

2. S-Lab and Microsoft develop Otter - a model proficient in multimodal perception and reasoning and MIMIC-IT - a large dataset for training vision-language models (Paper's website)

   

3. Researchers unveil a novel zero-shot text-gui…