haotian-liu / LLaVA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
https://llava.hliu.cc
Apache License 2.0
17.8k stars 1.92k forks source link

Inference errors when multiple questions #1393

Open samueleruffino99 opened 2 months ago

Z1zs commented 3 weeks ago

Hi Ruffino, it maybe somehow irrelevant to your question, but I'm really curious how did you construct the Autonomous Driving Scene Discussion. It looks so great. Did you build it by yourself or borrow it from some existing papers? Thank you very much!!!