Table 11 Time performance on LIAR dataset.

From: Graph-augmented transformer ensemble framework for robust and scalable fake news detection in social media ecosystems

Model

Training Time/Epoch (s)

Total Training Time (h)

Inference Time/Doc (ms)

GNN Only

18

1.5

2.1

Transformer Only (BERT)

42

3.2

4.8

Transformer + GNN

55

4.0

6.2

Full GETE Ensemble

68

4.8

7.5

Back to article page