# X-WAD: eXplainable Web Anomaly Detection

arXiv: https://arxiv.org/abs/2608.27172 · HTML: https://mlsanity.com/paper/2608.27172/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security.
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Placed in

- LLM safety and security / [Backdoor and data poisoning attacks](https://mlsanity.com/safety/backdoor-and-data-poisoning-attacks/)

## Builds on (most cited first, 12 of 12)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference](https://mlsanity.com/paper/2412.13663/) (2024)
- [SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model](https://mlsanity.com/paper/2502.02737/) (2025)
- [Problems with Shapley-value-based explanations as feature importance measures](https://mlsanity.com/paper/2002.11097/) (2020)
- [LogBERT: Log Anomaly Detection via BERT](https://mlsanity.com/paper/2103.04475/) (2021)
- [Interpretable Online Log Analysis Using Large Language Models with Prompt Strategies](https://mlsanity.com/paper/2308.07610/) (2023)
- [The Inadequacy of Shapley Values for Explainability](https://mlsanity.com/paper/2302.08160/) (2023)
- [On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMs](https://mlsanity.com/paper/2407.19200/) (2024)
- [HTTP2vec: Embedding of HTTP Requests for Detection of Anomalous Traffic](https://mlsanity.com/paper/2108.01763/) (2021)
- [Capturing the security expert knowledge in feature selection for web application attack detection](https://mlsanity.com/paper/2407.18445/) (2024)
- [ADALog: Adaptive Unsupervised Anomaly detection in Logs with Self-attention Masked Language Model](https://mlsanity.com/paper/2505.13496/) (2025)

## Abstract (arXiv)

The rapid growth of web-based services, particularly API-driven architectures, reflects an increasing reliance on distributed systems, exposing sensitive data to security risks and making the adoption of automated defensive mechanisms essential. In this context, where benign traffic predominates in real-world settings, modern defenses increasingly model normal behavior, relying on semi-supervised approaches trained on only normal data. However, ensuring the complete absence of anomalous instances in such training data is inherently difficult in practice, and mislabeled or contaminated attack samples can introduce backdoors into the learned defense, causing the model to silently misclassify certain attack patterns as normal behavior. This paper investigates the effectiveness of Transformer-based Language Models (TLMs) in the detection of anomalies in HTTP requests, focussing on providing detailed explanations for the detected anomalies. The study employs token-level logit-based surprisal mapping to provide both an anomaly score and a direct, detailed explanation via heatmap-like highlighting. The effectiveness of the proposed explainability approach is demonstrated by the discovery of labelling inconsistencies in a popular public dataset, revealing how anomalous contamination in the training data had induced backdoor-like failures in the detection models.