Table 3. The performace of our method and other state-of-art methods on DFDC dataset.

Method	ACC	AUC (%)
Xception (Chollet, 2017)	89.83	94.86
${\mathrm{F}}^3$Net (Qian et al., 2020)	79.86	87.50
M2TR (Wang et al., 2022)	91.27	97.20
GocNet (Guo et al., 2023c)	92.52	96.87
LDFnet (Guo et al., 2023b)	93.15	97.20
Our	94.03	98.57

Note:

Bold values refer to the best values.