IFM: Integrating and fine-tuning adversarial examples of recommendation system under multiple models to enhance their transferability

Fulan Qian, Yan Cui, Mengyao Xu, Hai Chen, Wenbin Chen, Qian Xu, Caihong Wu, Yuanting Yan, Shu Zhao

Abstract: Highlights•IFM integrates attack features from multiple surrogate models to enhance transferability.•IFM reduces overfitting of adversarial examples and improves transferability.•IFM performs well across different datasets and recommendation models.