EAGER: Edge-Aligned LLM Defense for Robust, Efficient, and Accurate Cybersecurity Question Answering
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918217360867328 |
|---|---|
| author | Gungor, Onat Sood, Roshan Zhou, Jiasheng Rosing, Tajana |
| author_facet | Gungor, Onat Sood, Roshan Zhou, Jiasheng Rosing, Tajana |
| contents | Large Language Models (LLMs) are highly effective for cybersecurity question answering (QA) but are difficult to deploy on edge devices due to their size. Quantization reduces memory and compute requirements but often degrades accuracy and increases vulnerability to adversarial attacks. We present EAGER, an edge-aligned defense framework that integrates parameter-efficient quantization with domain-specific preference alignment to jointly optimize efficiency, robustness, and accuracy. Unlike prior methods that address these aspects separately, EAGER leverages Quantized Low-Rank Adaptation (QLoRA) for low-cost fine-tuning and Direct Preference Optimization (DPO) on a self-constructed cybersecurity preference dataset, eliminating the need for human labels. Experiments show that EAGER reduces adversarial attack success rates by up to 7.3x and improves QA accuracy by up to 55% over state-of-the-art defenses, while achieving the lowest response latency on a Jetson Orin, demonstrating its practical edge deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_19523 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | EAGER: Edge-Aligned LLM Defense for Robust, Efficient, and Accurate Cybersecurity Question Answering Gungor, Onat Sood, Roshan Zhou, Jiasheng Rosing, Tajana Cryptography and Security Large Language Models (LLMs) are highly effective for cybersecurity question answering (QA) but are difficult to deploy on edge devices due to their size. Quantization reduces memory and compute requirements but often degrades accuracy and increases vulnerability to adversarial attacks. We present EAGER, an edge-aligned defense framework that integrates parameter-efficient quantization with domain-specific preference alignment to jointly optimize efficiency, robustness, and accuracy. Unlike prior methods that address these aspects separately, EAGER leverages Quantized Low-Rank Adaptation (QLoRA) for low-cost fine-tuning and Direct Preference Optimization (DPO) on a self-constructed cybersecurity preference dataset, eliminating the need for human labels. Experiments show that EAGER reduces adversarial attack success rates by up to 7.3x and improves QA accuracy by up to 55% over state-of-the-art defenses, while achieving the lowest response latency on a Jetson Orin, demonstrating its practical edge deployment. |
| title | EAGER: Edge-Aligned LLM Defense for Robust, Efficient, and Accurate Cybersecurity Question Answering |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2511.19523 |