[NeurIPS'23 Oral] Visual Instruction Tuning: LLaVA (Large Language-and-Vision Assistant) built towards GPT-4V level capabilities.
By chatting or signing in you agree to the Terms and chat-message logging (revocable in History).