Revisiting Image Captioning Training Paradigm via Direct CLIP-based Optimization
Paper
•
2408.14547
•
Published
Nicholas Moratelli
NicholasMoratelli
AI & ML interests
Multimodal Large Language Models - Vision and Language - Foundation Models - GenAI - Compositional AI
Recent Activity
authored
a paper
about 1 month ago
Revisiting Image Captioning Training Paradigm via Direct CLIP-based
Optimization
Organizations
NicholasMoratelli's activity
authored
a
paper
about 1 month ago