Implementation for the paper "Unified Multimodal Model with Unlikelihood Training for Visual Dialog" - View it on GitHub