ChunyuanLI

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Instruction Tuning with GPT-4

Grounded Language-Image Pre-training

Large Language-and-Vision Assistant for Biomedicine, built towards multimodal GPT-4 level capabilities.

Open-Set Grounded Text-to-Image Generation

Oscar and VinVL

LLaVA-Plus: Large Language and Vision Assistants that Plug and Learn to Use Skills