Adversarial-resilient lightweight phishing url detection: Evaluating lexical & metadata features under evasion techniques

Crafting evasive phishing URLs: exploiting tokenizer vulnerabilities in transformer-based detection systems

Phishing GAT: Adversarial-Hardened Phishing Email Detection via Semantic-Structural Fusion and Graph Attention Networks

Phishing email detection has been destabilised by Large Language Models (LLMs), which allow attackers to mass-produce fluent, contextually plausible messages. Detectors tuned on legacy corpora, notably the 2008 CEAS spam collection, consequently report benchmark accuracies that do not survive deployment. We present PhishingGAT, a detector that fuses word-level semantic features with structural ones and is hardened against adversarial perturbation. Each email is represented as an undirected graph whose nodes are unique tokens carrying 100-dimensional GloVe embeddings, and whose edges record co- occurrence inside a three-token sliding window. A two-layer Graph Attention Network (GAT) then learns per-edge importance over that graph. Robustness comes from Projected Gradient Descent (PGD) adversarial training, applied to continuous node features during optimisation only. Training uses a hybrid corpus of 70,716 emails assembled from the CEAS collection, the Enron collection, and curated LLM-generated samples. Evaluation follows two protocols. The first is strictly zero-shot: 1,186 unseen LLM-generated adversarial emails, on which the model reaches 77.40 per cent accuracy and an area under the receiver operating characteristic curve (AUC) of 0.8700. The second is domain-adapted: 401 held-out hard samples, on which accuracy rises to 90.27 per cent and AUC to 0.9717. The 12.87 percentage-point difference between the two is, to our knowledge, the first such measurement reported for graph-based phishing detection, and it quantifies a degradation that same-distribution benchmarks systematically hide. An ablation separates the contribution of attention from that of adversarial defence, and repeated runs across independent random seeds confirm that the reported figures are reproducible rather than seed artefacts.

GenAI-augmented ensemble learning framework for phishing URL detection using lexical, host-based and sequential features

Experimental results show that the proposed stacking ensemble consistently achieves the best overall performance, while a moderate augmentation ratio of 25% provides the highest robustness under temporal and cross-dataset evaluation; higher augmentation levels reduce performance.

Large Language Models for phishing URL detection: A comparative study of LLaMA-3, GEMMA-7B, and traditional Machine Learning approaches

Phishing attacks remain a major cybersecurity threat, causing financial loss, data breaches, and reputational damage to individuals and organizations. Conventional detection methods, such as rule-based filters and blacklist-based systems, often fail to identify newly generated or highly obfuscated phishing URLs. Recent advances in artificial intelligence, particularly large language models (LLMs), provide new opportunities to detect complex structural and contextual patterns in malicious web addresses. This study evaluates the effectiveness of transformer-based LLMs for phishing URL detection by comparing LLaMA-3 and GEMMA-7B with traditional machine learning models, including Support Vector Machines, Random Forest, and XGBoost. The models were fine-tuned and tested on a publicly available dataset containing 247,950 labeled URLs, each described by 42 structural and lexical features. To support transformer-based classification, structured URL attributes were converted into textual representations, enabling the models to exploit contextual embeddings. Experimental results showed that LLMs outperformed conventional machine learning baselines. GEMMA-7B achieved the best performance, with an accuracy of approximately 94.8%, followed by LLaMA-3 at 94%. These findings suggest that transformer-based models can effectively identify deceptive domain structures, abnormal URL patterns, and obfuscation techniques. Overall, the proposed LLM-based framework offers a promising approach for improving phishing detection and strengthening modern cybersecurity defenses.

Adversarial Robustness in Smishing Detection: A Comparative Analysis of Adversarial Fragility in Classical vs. Transformer-Based Detection Systems

Smishing detection systems are commonly trained and evaluated on clean, monolingual text. In low-resource settings, however, attackers frequently circumvent these systems through character obfuscation, cross-lingual code-switching, and structural perturbation. This study evaluates adversarial robustness for five model architectures: three classical lexical models (Random Forest, XGBoost, CNN+BiLSTM) and two multilingual transformers (mBERT, XLM-RoBERTa), using a dataset of 27,037 messages. Classical models are subjected to black-box generic attacks, while transformers are evaluated with attention-guided targeting. Each model is tested across three attack types and intensity levels, with performance measured by the Robustness Degradation Ratio (RDR). The results reveal a distinct architectural boundary: classical models experience near-catastrophic failure under character obfuscation and structural perturbation (RDR up to 0.988), whereas transformers demonstrate significantly greater resilience (RDR up to 0.351), with structural perturbation representing their most pronounced vulnerability. Effect-size analysis (Cliff's d) indicates a substantial difference between the two model categories. Within the transformer group, XLM-RoBERTa, despite achieving a higher clean-text baseline, exhibits greater degradation than mBERT. These findings demonstrate that clean-text performance is not a reliable predictor of adversarial robustness. Statistical validation using Mann-Whitney U and Friedman tests confirms that these patterns are attributable to model architecture rather than sampling. The results underscore the necessity for architecture-specific defences and frame smishing detection as an adversarial cybersecurity challenge rather than a static classification task.

Browser-based phishing detection system using modern web technologies

Phishing remains one of the most persistent cybersecurity threats, exploiting human trust to steal sensitive information through deceptive websites. Traditional detection methods, reliant on blacklists and reactive reporting, offer limited protection against rapidly evolving zero-day attacks. To address these challenges, this study proposes and evaluates a hybrid browser-resident phishing detection framework that combines three complementary detection mechanisms: a locally executed Random Forest model using URL lexical features, lightweight real-time DOM structure analysis, and VirusTotal’s multi-engine reputation service. Communication between the client and server is optimised using gRPC over the QUIC protocol, ensuring secure, high-performance data exchange with built-in retry mechanisms for resilience. The framework adopts a fail-safe decision strategy in which URLs identified as malicious by VirusTotal are immediately blocked, while benign or unknown results are further evaluated using a weighted fusion of machine learning and DOM-based scores. The model was trained on several high-impact URL features to ensure both speed and accuracy. Privacy remains a core principle, with sensitive data processed locally and minimal external data transmission. Experimental evaluation demonstrates that the proposed framework achieves a detection accuracy of 96.7%, with 97.1% precision, 96.0% recall, an F1-score of 96.6%, and an AUC of 0.993 using the optimised Random Forest classifier. The complete detection pipeline maintains an average end-to-end latency of less than one second, while the hybrid decision-fusion strategy successfully detects 91.7% of zero-day phishing URLs in the live evaluation dataset. These results demonstrate that the proposed architecture provides an accurate, low-latency, privacy-aware, and practically deployable solution for real-time browser-based phishing detection.