/VoCoT

VoCoT: Unleashing Visually Grounded Multi-Step Reasoning in Large Multi-Modal Models

Primary LanguagePython

Watchers