From ML to LLM: Evaluating the Robustness of Phishing Webpage Detection Models against Adversarial Attacks

TL;DR

This paper introduces PhishOracle, a tool for generating diverse adversarial phishing webpages, and evaluates the robustness of existing detection models, revealing significant vulnerabilities and the need for more resilient solutions.

Contribution

We develop PhishOracle to create diverse adversarial phishing webpages and evaluate the robustness of current detection models, highlighting their vulnerabilities against sophisticated attacks.

Findings

Detection models show significant drop in accuracy against adversarial webpages

Multimodal large language model-based detectors are more robust but still vulnerable

Many adversarial phishing webpages can deceive both models and users

Abstract

Phishing attacks attempt to deceive users into stealing sensitive information, posing a significant cybersecurity threat. Advances in machine learning (ML) and deep learning (DL) have led to the development of numerous phishing webpage detection solutions, but these models remain vulnerable to adversarial attacks. Evaluating their robustness against adversarial phishing webpages is essential. Existing tools contain datasets of pre-designed phishing webpages for a limited number of brands, and lack diversity in phishing features. To address these challenges, we develop PhishOracle, a tool that generates adversarial phishing webpages by embedding diverse phishing features into legitimate webpages. We evaluate the robustness of three existing task-specific models - Stack model, VisualPhishNet, and Phishpedia - against PhishOracle-generated adversarial phishing webpages and observe a significant drop in their detection rates. In contrast, a multimodal large language model (MLLM)-based phishing detector demonstrates stronger robustness against these adversarial attacks but still is prone to evasion. Our findings highlight the vulnerability of phishing detection models to adversarial attacks, emphasizing the need for more robust detection approaches. Furthermore, we conduct a user study to evaluate whether PhishOracle-generated adversarial phishing webpages can deceive users. The results show that many of these phishing webpages evade not only existing detection models but also users.