mlsanityData to 14 Oct 2026

EAGER: Edge-Aligned LLM Defense for Robust, Efficient, and Accurate Cybersecurity Question Answering

arXiv 2511.19523 ↗First posted 24 Nov 2025cs.CRcited by 2

On the map: LLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks

Abstract (arXiv)

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.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 19 of its 43 references that are in the map. First posted on arXiv 24 Nov 2025.

Builds on

19 of 43 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024SecAlign: Defending Against Prompt Injection with Preference Optimization7 Oct

Other references · 18

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  4. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,418
  5. 2024StruQ: Defending Against Prompt Injection with Structured Queriescited by 375
  6. 2024Instructional Segment Embedding: Improving LLM Safety with Instruction Hierarchycited by 67
  7. 2024Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compressioncited by 57
  8. 2024Defense Against Prompt Injection Attack by Leveraging Attack Techniquescited by 40
  9. 2023SecQA: A Concise Question-Answering Dataset for Evaluating Large Language Models in Computer Securitycited by 37
  10. 2023Agile-Quant: Activation-Guided Quantization for Faster Inference of LLMs on the Edgecited by 33
  11. 2025Llama-3.1-FoundationAI-SecurityLLM-Base-8B Technical Reportcited by 31
  12. 2024Fine-Tuning, Quantization, and LLMs: Navigating Unintended Outcomescited by 14
6 more
  1. 2024Investigating the Impact of Quantization on Adversarial Robustnesscited by 9
  2. 2024Improving Conversational Abilities of Quantized Large Language Models via Direct Preference Alignmentcited by 9
  3. 2024A RAG-Based Question-Answering Solution for Cyber-Attack Investigation and Attributioncited by 9
  4. 2024Edge Intelligence Optimization for Large Language Model Inference with Batching and Quantizationcited by 7
  5. 2025AQUA-LLM: Evaluating Accuracy, Quantization, and Adversarial Robustness Trade-offs in LLMs for Cybersecurity Question Answeringcited by 3
  6. 2023A Threat-Intelligence Driven Methodology to Incorporate Uncertainty in Cyber Risk Analysis and Enhance Decision Makingcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026COPA: Continual Preference Optimization for Adaptive Prompt Injection Defensecited by 2
  2. 2026CyberMaskQA: A Privacy-Aware Benchmark for Evaluating Large Language Models in Cybersecurity Question Answeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Fine-tuning of Large Language Models for Domain-Specific Cybersecurity Knowledge
  2. 2024AttackQA: Development and Adoption of a Dataset for Assisting Cybersecurity Operations using Fine-tuned and Open-Source LLMs
  3. 2024Safety Layers in Aligned Large Language Models: The Key to LLM Security
  4. 2025REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective
  5. 2026A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation
  6. 2025Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs
  7. 2025Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives
  8. 2026Suan: Rectifying Direct Preference Safety Alignment in Large Language Models