VisualScratchpad: Grounding Visual Concepts in Large Vision Language Models
Grounding visual concepts in large vision-language models via a attention-based linking mechanism.
hyesu-lim
Dongyoon Han
•
•
1 min read
Towards Calibrated Robust Fine-Tuning of Vision-Language Models
Calibrated, robust fine-tuning method for vision-language models that preserves uncertainty estimates under distribution shift.
changdae-oh