Fetching the paper…

PRADA: Practical Black-Box Adversarial Attacks against Neural Ranking Models · Around