Representation Bending for Large Language Model Safety

• URL: http://arxiv.org/abs/2504.01550v1
• Date: Wed, 02 Apr 2025 09:47:01 GMT
• Title: Representation Bending for Large Language Model Safety
• Authors: Ashkan Yousefpour, Taeheon Kim, Ryan S. Kwon, Seungbeen Lee, Wonje Jeung, Seungju Han, Alvin Wan, Harrison Ngan, Youngjae Yu, Jonghyun Choi,
• Abstract summary: Large Language Models (LLMs) have emerged as powerful tools, but their inherent safety risks pose significant challenges.<n>This paper introduces RepBend, a novel approach that fundamentally disrupts the representations underlying harmful behaviors in LLMs.<n>RepBend achieves state-of-the-art performance, outperforming prior methods such as Circuit Breaker, RMU, and NPO, with up to 95% reduction in attack success rates.
• Score: 27.842146980762934
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have emerged as powerful tools, but their inherent safety risks - ranging from harmful content generation to broader societal harms - pose significant challenges. These risks can be amplified by the recent adversarial attacks, fine-tuning vulnerabilities, and the increasing deployment of LLMs in high-stakes environments. Existing safety-enhancing techniques, such as fine-tuning with human feedback or adversarial training, are still vulnerable as they address specific threats and often fail to generalize across unseen attacks, or require manual system-level defenses. This paper introduces RepBend, a novel approach that fundamentally disrupts the representations underlying harmful behaviors in LLMs, offering a scalable solution to enhance (potentially inherent) safety. RepBend brings the idea of activation steering - simple vector arithmetic for steering model's behavior during inference - to loss-based fine-tuning. Through extensive evaluation, RepBend achieves state-of-the-art performance, outperforming prior methods such as Circuit Breaker, RMU, and NPO, with up to 95% reduction in attack success rates across diverse jailbreak benchmarks, all with negligible reduction in model usability and general capabilities.

Related papers

• Improving LLM Safety Alignment with Dual-Objective Optimization [65.41451412400609]
  Existing training-time safety alignment techniques for large language models (LLMs) remain vulnerable to jailbreak attacks. We propose an improved safety alignment that disentangles DPO objectives into two components: (1) robust refusal training, which encourages refusal even when partial unsafe generations are produced, and (2) targeted unlearning of harmful knowledge.
  arXiv  Detail & Related papers  (2025-03-05T18:01:05Z)
• Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities [49.09703018511403]
  Large language model (LLM) risks and capabilities are increasingly being incorporated into AI risk management and governance frameworks.<n>Currently, most risk evaluations are conducted by designing inputs that elicit harmful behaviors from the system.<n>We propose evaluating LLMs with model tampering attacks which allow for modifications to latent activations or weights.
  arXiv  Detail & Related papers  (2025-02-03T18:59:16Z)
• Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense [44.01174462291761]
  Large Language Models (LLMs) have showcased remarkable capabilities across various domains. activation approximation has emerged as a promising avenue for pursuing inference efficiency. Despite achieving substantial speedups with minimal impact on utility, the safety implications of activation approximations remain unclear.
  arXiv  Detail & Related papers  (2025-02-02T16:25:48Z)
• Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning [21.423429565221383]
  Large language models (LLMs) are vital for a wide range of applications yet remain susceptible to jailbreak threats. We propose a novel defense strategy, Safety Chain-of-Thought (SCoT), which harnesses the enhanced textitreasoning capabilities of LLMs for proactive assessment of harmful inputs.
  arXiv  Detail & Related papers  (2025-01-31T14:45:23Z)
• Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks [34.40254709148148]
  Pre-trained vision-language models (VLMs) have showcased remarkable performance in image and natural language understanding. Their potential safety and robustness issues raise concerns that adversaries may evade the system and cause these models to generate toxic content through malicious attacks. We present Chain of Attack (CoA), which iteratively enhances the generation of adversarial examples based on the multi-modal semantic update.
  arXiv  Detail & Related papers  (2024-11-24T05:28:07Z)
• Transferable Adversarial Attacks on SAM and Its Downstream Models [87.23908485521439]
  This paper explores the feasibility of adversarial attacking various downstream models fine-tuned from the segment anything model (SAM) To enhance the effectiveness of the adversarial attack towards models fine-tuned on unknown datasets, we propose a universal meta-initialization (UMI) algorithm.
  arXiv  Detail & Related papers  (2024-10-26T15:04:04Z)
• Root Defence Strategies: Ensuring Safety of LLM at the Decoding Level [10.476222570886483]
  Large language models (LLMs) have demonstrated immense utility across various industries.<n>As LLMs advance, the risk of harmful outputs increases due to incorrect or malicious instruction prompts.<n>This paper examines the LLMs' capability to recognize harmful outputs, revealing and quantifying their proficiency in assessing the danger of previous tokens.
  arXiv  Detail & Related papers  (2024-10-09T12:09:30Z)
• SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance [48.36220909956064]
  SafeAligner is a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks.<n>We develop two specialized models: the Sentinel Model, which is trained to foster safety, and the Intruder Model, designed to generate riskier responses.<n>We show that SafeAligner can increase the likelihood of beneficial tokens, while reducing the occurrence of harmful ones.
  arXiv  Detail & Related papers  (2024-06-26T07:15:44Z)
• RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content [62.685566387625975]
  Current mitigation strategies, while effective, are not resilient under adversarial attacks. This paper introduces Resilient Guardrails for Large Language Models (RigorLLM), a novel framework designed to efficiently moderate harmful and unsafe inputs.
  arXiv  Detail & Related papers  (2024-03-19T07:25:02Z)
• Uniformly Safe RL with Objective Suppression for Multi-Constraint Safety-Critical Applications [73.58451824894568]
  The widely adopted CMDP model constrains the risks in expectation, which makes room for dangerous behaviors in long-tail states. In safety-critical domains, such behaviors could lead to disastrous outcomes. We propose Objective Suppression, a novel method that adaptively suppresses the task reward maximizing objectives according to a safety critic.
  arXiv  Detail & Related papers  (2024-02-23T23:22:06Z)
• Visual Adversarial Examples Jailbreak Aligned Large Language Models [66.53468356460365]
  We show that the continuous and high-dimensional nature of the visual input makes it a weak link against adversarial attacks. We exploit visual adversarial examples to circumvent the safety guardrail of aligned LLMs with integrated vision. Our study underscores the escalating adversarial risks associated with the pursuit of multimodality.
  arXiv  Detail & Related papers  (2023-06-22T22:13:03Z)

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.