Image Captioning on Flickr30K Translated Portuguese (test)
62.03CIDEr-DViTucano 2B
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| ViTucano 2B#Params=2.88B, Evaluation Mode=Zero-shot2026.01 | 62.03 | 19.84 | 41.51 | 47.05 | 70.3 | 56.28 | |
| ViTucano 1B#Params=1.53B, Evaluation Mode=Zero-shot2026.01 | 57.96 | 19.47 | 40.8 | 47.01 | 70.09 | 55.96 | |
| LLaMa3.2 V#Params=11.70B, Evaluation Mode=Zero-shot2026.01 | 34.48 | 10.2 | 31.56 | 35.17 | 63.18 | 56.95 | |
| GPT-4oEvaluation Mode=Zero-shot2026.01 | 25.68 | 7.36 | 25.04 | 38.67 | 61.49 | 62.07 | |
| PaliGemma#Params=2.92B, Evaluation Mode=Zero-shot2026.01 | 23.47 | 4.88 | 20.76 | 20.33 | 52.37 | 49.67 | |
| Phi-3 V#Params=4.15B, Evaluation Mode=Zero-shot2026.01 | 22.67 | 7.17 | 27.35 | 29.5 | 58.44 | 52.36 | |
| GPT-4o-miniEvaluation Mode=Zero-shot2026.01 | 21.68 | 8.96 | 29.31 | 41.63 | 63.83 | 61.26 |