Table 10 Study of W2VV components: impact of model configurations on sentence recommendation performance in Arabic.
Model configuration | BLEU-4 | METEOR | ROUGE-L | CIDEr | SPICE |
|---|---|---|---|---|---|
Baseline MLP | 67.0 | 58.0 | 65.0 | 120.0 | 43.5 |
W2VV | 70.0 | 61.0 | 67.0 | 125.0 | 45.0 |
W2VV + Attention +  Contrastive Loss | 74.1 | 64.6 | 70.4 | 133.2 | 47.5 |
