Image Captioning on normal
73.67BLEU-1VIT + GPT (SOLI-par)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VIT + GPT (SOLI-par)Architecture=VIT + GPT, SOLI approach=SOLI-par2025.12 | 73.67 | 64.7 | 56.93 | |
| VIT + GPT (SOLI-half)Architecture=VIT + GPT, SOLI approach=SOLI-half2025.12 | 72.06 | 63.12 | 56.1 | |
| ResNet+Att-LSTM-GloVe (SOLI-par)Architecture=ResNet+Att-LSTM-GloVe, SOLI approach=SOLI-par2025.12 | 59.66 | 21.87 | 24.36 | |
| ResNet+Att-LSTM-GloVe (SOLI-half)Architecture=ResNet+Att-LSTM-GloVe, SOLI approach=SOLI-half2025.12 | 58.25 | 20.38 | 22.28 |