Jina-vlm delivers state-of-the-art multilingual VQA performance among open 2B-scale vision-language models through efficient image handling and systematic data ablation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
jina-vlm: Small Multilingual Vision Language Model
Jina-vlm delivers state-of-the-art multilingual VQA performance among open 2B-scale vision-language models through efficient image handling and systematic data ablation.