Fine-tuned multimodal LLMs auto-generates rich cultural-heritage descriptions for visually impaired audiences in Romanian, Serbian and German.
Iosif Tsangko
tsangkos
·
AI & ML interests
None yet
Organizations
None yet
models 11
tsangkos/SmolVLM-500M-Instruct-fer0
Image-Text-to-Text • 0.5B • Updated • 61 • 2
tsangkos/ArtQwen-Curator-RO
Image-to-Text • Updated
tsangkos/ArtQwen-Curator-SR
Image-to-Text • Updated
tsangkos/ArtQwen-Curator-DE
Image-to-Text • Updated
tsangkos/ArtQwen-Curator-ML
Image-to-Text • Updated
tsangkos/gemma-2-2B-it-thinking-function_calling-shift-V1
Updated • 5
tsangkos/OSCAR-ro-llama7
Updated
tsangkos/OSCAR-ro-llama1.3
Updated
tsangkos/OSCAR-ro-mistral
Updated
tsangkos/OSCAR-ro-Llama-3-8B-Instruct
Updated
datasets 0
None public yet