# Evaluating Large Language Models for Phishing Detection, Self-Consistency, Faithfulness, and Explainability

arXiv: https://arxiv.org/abs/2506.13746 · HTML: https://mlsanity.com/paper/2506.13746/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 14 of its 28 references that are in the map.
- First posted on arXiv 16 Jun 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 14 of 14)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [Keystroke Dynamics as Part of Lifelogging](https://mlsanity.com/paper/2010.10144/) (2020)
- [Data-efficient Fine-tuning for LLM-based Recommendation](https://mlsanity.com/paper/2401.17197/) (2024)
- [On Measuring Faithfulness or Self-consistency of Natural Language Explanations](https://mlsanity.com/paper/2311.07466/) (2023)
- [Spear Phishing With Large Language Models](https://mlsanity.com/paper/2305.06972/) (2023)
- [ChatSpamDetector: Leveraging Large Language Models for Effective Phishing Email Detection](https://mlsanity.com/paper/2402.18093/) (2024)
- [Next-Generation Phishing: How LLM Agents Empower Cyber Attackers](https://mlsanity.com/paper/2411.13874/) (2024)
- [An Explainable Transformer-based Model for Phishing Email Detection: A Large Language Model Approach](https://mlsanity.com/paper/2402.13871/) (2024)
- [Can Machines Think Like Humans? A Behavioral Evaluation of LLM Agents in Dictator Games](https://mlsanity.com/paper/2410.21359/) (2024)
- [Enhancing Phishing Email Identification with Large Language Models](https://mlsanity.com/paper/2502.04759/) (2025)
- [EXPLICATE: Enhancing Phishing Detection through Explainable AI and LLM-Powered Interpretability](https://mlsanity.com/paper/2503.20796/) (2025)

## Built on by (most cited first, 1 of 1)

- [Every Character Counts: From Vulnerability to Defense in Phishing Detection](https://mlsanity.com/paper/2509.20589/) (2025)

## Abstract (arXiv)

Phishing attacks remain one of the most prevalent and persistent cybersecurity threat with attackers continuously evolving and intensifying tactics to evade the general detection system. Despite significant advances in artificial intelligence and machine learning, faithfully reproducing the interpretable reasoning with classification and explainability that underpin phishing judgments remains challenging. Due to recent advancement in Natural Language Processing, Large Language Models (LLMs) show a promising direction and potential for improving domain specific phishing classification tasks. However, enhancing the reliability and robustness of classification models requires not only accurate predictions from LLMs but also consistent and trustworthy explanations aligning with those predictions. Therefore, a key question remains: can LLMs not only classify phishing emails accurately but also generate explanations that are reliably aligned with their predictions and internally self-consistent? To answer these questions, we have fine-tuned transformer based models, including BERT, Llama models, and Wizard, to improve domain relevance and make them more tailored to phishing specific distinctions, using Binary Sequence Classification, Contrastive Learning (CL) and Direct Preference Optimization (DPO). To that end, we examined their performance in phishing classification and explainability by applying the ConsistenCy measure based on SHAPley values (CC SHAP), which measures prediction explanation token alignment to test the model's internal faithfulness and consistency and uncover the rationale behind its predictions and reasoning. Overall, our findings show that Llama models exhibit stronger prediction explanation token alignment with higher CC SHAP scores despite lacking reliable decision making accuracy, whereas Wizard achieves better prediction accuracy but lower CC SHAP scores.