English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Neural ranking models (NRMs) have shown remarkable success in recent years,especially with pre-trained language models. However, deep neural models arenotorious for their vulnerability to adversarial examples. Adversarial attacksmay become a new type of web spamming technique given our increased reliance onneural information retrieval models. Therefore, it is important to studypotential adversarial attacks to identify vulnerabilities of NRMs before theyare deployed. In this paper, we introduce the Word Substitution Ranking Attack(WSRA) task against NRMs, which aims to promote a target document in rankingsby adding adversarial perturbations to its text. We focus on the decision-basedblack-box attack setting, where the attackers cannot directly get access to themodel information, but can only query the target model to obtain the rankpositions of the partial retrieved list. This attack setting is realistic inreal-world search engines. We propose a novel Pseudo Relevance-basedADversarial ranking Attack method (PRADA) that learns a surrogate model basedon Pseudo Relevance Feedback (PRF) to generate gradients for finding theadversarial perturbations. Experiments on two web search benchmark datasetsshow that PRADA can outperform existing attack strategies and successfully foolthe NRM with small indiscernible perturbations of text.

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