SHIFT: Vision-Language Models for Accessible CH
Collection
Fine-tuned multimodal LLMs auto-generates rich cultural-heritage descriptions for visually impaired audiences in Romanian, Serbian and German.
•
7 items
•
Updated
•
1
