Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities

• URL: http://arxiv.org/abs/2502.05209v1
• Date: Mon, 03 Feb 2025 18:59:16 GMT
• Title: Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities
• Authors: Zora Che, Stephen Casper, Robert Kirk, Anirudh Satheesh, Stewart Slocum, Lev E McKinney, Rohit Gandikota, Aidan Ewart, Domenic Rosati, Zichu Wu, Zikui Cai, Bilal Chughtai, Yarin Gal, Furong Huang, Dylan Hadfield-Menell,
• Abstract summary: Large language model (LLM) risks and capabilities are increasingly being incorporated into AI risk management and governance frameworks. Currently, most risk evaluations are conducted by designing inputs that elicit harmful behaviors from the system. We propose evaluating LLMs with model tampering attacks which allow for modifications to latent activations or weights.
• Score: 49.09703018511403
• License:
• Abstract: Evaluations of large language model (LLM) risks and capabilities are increasingly being incorporated into AI risk management and governance frameworks. Currently, most risk evaluations are conducted by designing inputs that elicit harmful behaviors from the system. However, a fundamental limitation of this approach is that the harmfulness of the behaviors identified during any particular evaluation can only lower bound the model's worst-possible-case behavior. As a complementary method for eliciting harmful behaviors, we propose evaluating LLMs with model tampering attacks which allow for modifications to latent activations or weights. We pit state-of-the-art techniques for removing harmful LLM capabilities against a suite of 5 input-space and 6 model tampering attacks. In addition to benchmarking these methods against each other, we show that (1) model resilience to capability elicitation attacks lies on a low-dimensional robustness subspace; (2) the attack success rate of model tampering attacks can empirically predict and offer conservative estimates for the success of held-out input-space attacks; and (3) state-of-the-art unlearning methods can easily be undone within 16 steps of fine-tuning. Together these results highlight the difficulty of removing harmful LLM capabilities and show that model tampering attacks enable substantially more rigorous evaluations than input-space attacks alone. We release models at https://huggingface.co/LLM-GAT

Related papers

• On Evaluating the Durability of Safeguards for Open-Weight LLMs [80.36750298080275]
  We discuss whether technical safeguards can impede the misuse of large language models (LLMs) We show that even evaluating these defenses is exceedingly difficult and can easily mislead audiences into thinking that safeguards are more durable than they really are. We suggest future research carefully cabin claims to more constrained, well-defined, and rigorously examined threat models.
  arXiv  Detail & Related papers  (2024-12-10T01:30:32Z)
• 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. As LLMs advance, the risk of harmful outputs increases due to incorrect or malicious instruction prompts. 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)
• 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)
• Self-Evaluation as a Defense Against Adversarial Attacks on LLMs [20.79833694266861]
  We introduce a defense against adversarial attacks on LLMs utilizing self-evaluation. Our method requires no model fine-tuning, instead using pre-trained models to evaluate the inputs and outputs of a generator model. We present an analysis of the effectiveness of our method, including attempts to attack the evaluator in various settings.
  arXiv  Detail & Related papers  (2024-07-03T16:03:42Z)
• Defending Large Language Models Against Attacks With Residual Stream Activation Analysis [0.0]
  Large Language Models (LLMs) are vulnerable to adversarial threats. This paper presents an innovative defensive strategy, given white box access to an LLM. We apply a novel methodology for analyzing distinctive activation patterns in the residual streams for attack prompt classification.
  arXiv  Detail & Related papers  (2024-06-05T13:06:33Z)
• Exploiting LLM Quantization [6.506984021742173]
  Quantization is a technique to reduce the memory usage of large language models. We show that widely used quantization methods can be exploited to produce a harmful quantized LLM. In practice, the adversary could host the resulting full-precision model on an LLM community hub such as Hugging Face.
  arXiv  Detail & Related papers  (2024-05-28T12:51:01Z)
• Unveiling the Misuse Potential of Base Large Language Models via In-Context Learning [61.2224355547598]
  Open-sourcing of large language models (LLMs) accelerates application development, innovation, and scientific progress. Our investigation exposes a critical oversight in this belief. By deploying carefully designed demonstrations, our research demonstrates that base LLMs could effectively interpret and execute malicious instructions.
  arXiv  Detail & Related papers  (2024-04-16T13:22:54Z)
• ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models [64.03398193325572]
  Inference attacks against Machine Learning (ML) models allow adversaries to learn about training data, model parameters, etc. We concentrate on four attacks - namely, membership inference, model inversion, attribute inference, and model stealing. Our analysis relies on a modular re-usable software, ML-Doctor, which enables ML model owners to assess the risks of deploying their models.
  arXiv  Detail & Related papers  (2021-02-04T11:35:13Z)

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.