# Scam Shield: Multi-Model Voting and Fine-Tuned LLMs Against Adversarial Attacks

arXiv: https://arxiv.org/abs/2511.01746 · HTML: https://mlsanity.com/paper/2511.01746/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 25 references that are in the map.
- First posted on arXiv 3 Nov 2025.

## Builds on (most cited first, 8 of 8)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers](https://mlsanity.com/paper/1801.04354/) (2018)
- [Data Augmentation Can Improve Robustness](https://mlsanity.com/paper/2111.05328/) (2021)
- [Med42 -- Evaluating Fine-Tuning Strategies for Medical LLMs: Full-Parameter vs. Parameter-Efficient Approaches](https://mlsanity.com/paper/2404.14779/) (2024)
- [Detecting Scams Using Large Language Models](https://mlsanity.com/paper/2402.03147/) (2024)
- [Combating Phone Scams with LLM-based Detection: Where Do We Stand?](https://mlsanity.com/paper/2409.11643/) (2024)
- [Exposing LLM Vulnerabilities: Adversarial Scam Detection and Performance](https://mlsanity.com/paper/2412.00621/) (2024)
- [An Empirical Analysis of SMS Scam Detection Systems](https://mlsanity.com/paper/2210.10451/) (2022)

## Abstract (arXiv)

Scam detection remains a critical challenge in cybersecurity as adversaries craft messages that evade automated filters. We propose a Hierarchical Scam Detection System (HSDS) that combines a lightweight multi-model voting front end with a fine-tuned LLaMA 3.1 8B Instruct back end to improve accuracy and robustness against adversarial attacks. An ensemble of four classifiers provides preliminary predictions through majority vote, and ambiguous cases are escalated to the fine-tuned model, which is optimized with adversarial training to reduce misclassification. Experiments show that this hierarchical design both improves adversarial scam detection and shortens inference time by routing most cases away from the LLM, outperforming traditional machine-learning baselines and proprietary LLM baselines. The findings highlight the effectiveness of a hybrid voting mechanism and adversarial fine-tuning in fortifying LLMs against evolving scam tactics, enhancing the resilience of automated scam detection systems.