Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data

• URL: http://arxiv.org/abs/2505.09974v1
• Date: Thu, 15 May 2025 05:22:53 GMT
• Title: Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data
• Authors: Adel ElZemity, Budi Arief, Shujun Li,
• Abstract summary: We present a systematic evaluation of safety risks in fine-tuned large language models (LLMs) for cyber security applications.<n>Our evaluation shows that fine-tuning reduces safety resilience across all tested LLMs.<n>We propose and evaluate a safety alignment approach that carefully rewords instruction-response pairs to include explicit safety precautions and ethical considerations.
• Score: 2.2530496464901106
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: The integration of large language models (LLMs) into cyber security applications presents significant opportunities, such as enhancing threat analysis and malware detection, but can also introduce critical risks and safety concerns, including personal data leakage and automated generation of new malware. We present a systematic evaluation of safety risks in fine-tuned LLMs for cyber security applications. Using the OWASP Top 10 for LLM Applications framework, we assessed seven open-source LLMs: Phi 3 Mini 3.8B, Mistral 7B, Qwen 2.5 7B, Llama 3 8B, Llama 3.1 8B, Gemma 2 9B, and Llama 2 70B. Our evaluation shows that fine-tuning reduces safety resilience across all tested LLMs (e.g., the safety score of Llama 3.1 8B against prompt injection drops from 0.95 to 0.15). We propose and evaluate a safety alignment approach that carefully rewords instruction-response pairs to include explicit safety precautions and ethical considerations. This approach demonstrates that it is possible to maintain or even improve model safety while preserving technical utility, offering a practical path forward for developing safer fine-tuning methodologies. This work offers a systematic evaluation for safety risks in LLMs, enabling safer adoption of generative AI in sensitive domains, and contributing towards the development of secure, trustworthy, and ethically aligned LLMs.

Related papers

• SafeLawBench: Towards Safe Alignment of Large Language Models [18.035407356604832]
  There is a lack of definitive standards for evaluating the safety of large language models (LLMs)<n>SafeLawBench categorizes safety risks into three levels based on legal standards.<n>It comprises 24,860 multi-choice questions and 1,106 open-domain question-answering (QA) tasks.
  arXiv  Detail & Related papers  (2025-06-07T03:09:59Z)
• SafeAgent: Safeguarding LLM Agents via an Automated Risk Simulator [77.86600052899156]
  Large Language Model (LLM)-based agents are increasingly deployed in real-world applications.<n>We propose AutoSafe, the first framework that systematically enhances agent safety through fully automated synthetic data generation.<n>We show that AutoSafe boosts safety scores by 45% on average and achieves a 28.91% improvement on real-world tasks.
  arXiv  Detail & Related papers  (2025-05-23T10:56:06Z)
• The Hidden Risks of LLM-Generated Web Application Code: A Security-Centric Evaluation of Code Generation Capabilities in Large Language Models [0.769672852567215]
  This paper uses predefined security parameters to evaluate the security compliance of LLM-generated code across multiple models.<n>The analysis reveals critical vulnerabilities in authentication mechanisms, session management, input validation and HTTP security headers.<n>Our findings underscore that human expertise is crucial to ensure secure software deployment or review of LLM-generated code.
  arXiv  Detail & Related papers  (2025-04-29T10:23:11Z)
• CyberLLMInstruct: A New Dataset for Analysing Safety of Fine-Tuned LLMs Using Cyber Security Data [2.2530496464901106]
  integration of large language models into cyber security applications presents significant opportunities.<n>CyberLLMInstruct is a dataset of 54,928 instruction-response pairs spanning cyber security tasks.<n>Fine-tuning models can achieve up to 92.50 percent accuracy on the CyberMetric benchmark.
  arXiv  Detail & Related papers  (2025-03-12T12:29:27Z)
• Internal Activation as the Polar Star for Steering Unsafe LLM Behavior [50.463399903987245]
  We introduce SafeSwitch, a framework that dynamically regulates unsafe outputs by monitoring and utilizing the model's internal states.<n>Our empirical results show that SafeSwitch reduces harmful outputs by over 80% on safety benchmarks while maintaining strong utility.
  arXiv  Detail & Related papers  (2025-02-03T04:23:33Z)
• Global Challenge for Safe and Secure LLMs Track 1 [57.08717321907755]
  The Global Challenge for Safe and Secure Large Language Models (LLMs) is a pioneering initiative organized by AI Singapore (AISG) and the CyberSG R&D Programme Office (CRPO) This paper introduces the Global Challenge for Safe and Secure Large Language Models (LLMs), a pioneering initiative organized by AI Singapore (AISG) and the CyberSG R&D Programme Office (CRPO) to foster the development of advanced defense mechanisms against automated jailbreaking attacks.
  arXiv  Detail & Related papers  (2024-11-21T08:20:31Z)
• SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models [75.67623347512368]
  We propose toolns, a comprehensive framework designed for conducting safety evaluations of MLLMs. Our framework consists of a comprehensive harmful query dataset and an automated evaluation protocol. Based on our framework, we conducted large-scale experiments on 15 widely-used open-source MLLMs and 6 commercial MLLMs.
  arXiv  Detail & Related papers  (2024-10-24T17:14:40Z)
• Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training [67.30423823744506]
  This study addresses a critical gap in safety tuning practices for Large Language Models (LLMs) We introduce a novel approach, Decoupled Refusal Training (DeRTa), designed to empower LLMs to refuse compliance to harmful prompts at any response position. DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful
  arXiv  Detail & Related papers  (2024-07-12T09:36:33Z)
• S-Eval: Towards Automated and Comprehensive Safety Evaluation for Large Language Models [46.148439517272024]
  Generative large language models (LLMs) have revolutionized natural language processing with their transformative and emergent capabilities.<n>Recent evidence indicates that LLMs can produce harmful content that violates social norms.<n>We propose S-Eval, an automated Safety Evaluation framework with a newly defined comprehensive risk taxonomy.
  arXiv  Detail & Related papers  (2024-05-23T05:34:31Z)
• Uncovering Safety Risks of Large Language Models through Concept Activation Vector [13.804245297233454]
  We introduce a Safety Concept Activation Vector (SCAV) framework to guide attacks on large language models (LLMs)<n>We then develop an SCAV-guided attack method that can generate both attack prompts and embedding-level attacks.<n>Our attack method significantly improves the attack success rate and response quality while requiring less training data.
  arXiv  Detail & Related papers  (2024-04-18T09:46:25Z)
• Safety Assessment of Chinese Large Language Models [51.83369778259149]
  Large language models (LLMs) may generate insulting and discriminatory content, reflect incorrect social values, and may be used for malicious purposes. To promote the deployment of safe, responsible, and ethical AI, we release SafetyPrompts including 100k augmented prompts and responses by LLMs.
  arXiv  Detail & Related papers  (2023-04-20T16:27:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.