Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MAWSEO: Adversarial Wiki Search Poisoning for Illicit Online Promotion

Zilong Lin, Zhengyi Li, Xiaojing Liao, XiaoFeng Wang, Xiaozhong Liu

TL;DR

This work demonstrates that Wiki search poisoning for illicit promotion can be automated as a black-box adversarial attack. It introduces MAWSEO, a two-phase system that first generates Wiki-style promotional paragraphs and then retrieves and inserts them into target articles using a multi-task adversarial passage retrieval framework with four discriminators. Empirical results on a local Wiki platform show MAWSEO achieving substantial rank boosts and evasion of vandalism detectors, while user studies indicate promotional content can reach readers without triggering alarm. The paper also proposes defenses—coherence-based detection and adversarial training—highlighting a path toward more robust Wiki ecosystems and outlining limitations and future directions.

Abstract

As a prominent instance of vandalism edits, Wiki search poisoning for illicit promotion is a cybercrime in which the adversary aims at editing Wiki articles to promote illicit businesses through Wiki search results of relevant queries. In this paper, we report a study that, for the first time, shows that such stealthy blackhat SEO on Wiki can be automated. Our technique, called MAWSEO, employs adversarial revisions to achieve real-world cybercriminal objectives, including rank boosting, vandalism detection evasion, topic relevancy, semantic consistency, user awareness (but not alarming) of promotional content, etc. Our evaluation and user study demonstrate that MAWSEO is capable of effectively and efficiently generating adversarial vandalism edits, which can bypass state-of-the-art built-in Wiki vandalism detectors, and also get promotional content through to Wiki users without triggering their alarms. In addition, we investigated potential defense, including coherence based detection and adversarial training of vandalism detection, against our attack in the Wiki ecosystem.

MAWSEO: Adversarial Wiki Search Poisoning for Illicit Online Promotion

TL;DR

This work demonstrates that Wiki search poisoning for illicit promotion can be automated as a black-box adversarial attack. It introduces MAWSEO, a two-phase system that first generates Wiki-style promotional paragraphs and then retrieves and inserts them into target articles using a multi-task adversarial passage retrieval framework with four discriminators. Empirical results on a local Wiki platform show MAWSEO achieving substantial rank boosts and evasion of vandalism detectors, while user studies indicate promotional content can reach readers without triggering alarm. The paper also proposes defenses—coherence-based detection and adversarial training—highlighting a path toward more robust Wiki ecosystems and outlining limitations and future directions.

Abstract

As a prominent instance of vandalism edits, Wiki search poisoning for illicit promotion is a cybercrime in which the adversary aims at editing Wiki articles to promote illicit businesses through Wiki search results of relevant queries. In this paper, we report a study that, for the first time, shows that such stealthy blackhat SEO on Wiki can be automated. Our technique, called MAWSEO, employs adversarial revisions to achieve real-world cybercriminal objectives, including rank boosting, vandalism detection evasion, topic relevancy, semantic consistency, user awareness (but not alarming) of promotional content, etc. Our evaluation and user study demonstrate that MAWSEO is capable of effectively and efficiently generating adversarial vandalism edits, which can bypass state-of-the-art built-in Wiki vandalism detectors, and also get promotional content through to Wiki users without triggering their alarms. In addition, we investigated potential defense, including coherence based detection and adversarial training of vandalism detection, against our attack in the Wiki ecosystem.
Paper Structure (26 sections, 10 equations, 4 figures, 9 tables)

This paper contains 26 sections, 10 equations, 4 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Wiki Systems
  4. Illicit Promotion on IR Systems
  5. Threat Model
  6. Ethical Consideration
  7. Methodologies
  8. Overview
  9. Data Preparation: Generation of Candidate Promotion Paragraphs
  10. Retrieval & Insertion of Promotion Paragraph
  11. Impelmentation and Evaluation
  12. Implementation
  13. Experiment Setup
  14. End-to-end Effectiveness and Efficiency
  15. User Reachability and Evasiveness
  16. ...and 11 more sections

Figures (4)

  • Figure 1: Overview of Wiki search poisoning for illicit promotion by MAWSEO. The attack includes a total of five steps across two phases.
  • Figure 2: Architecture of incentive injection model.
  • Figure 3: Architecture of multi-task adversarial passage retrieval model.
  • Figure 4: Coherence Detection Model.