The paper examines the vulnerability of open‑weight large language model safeguards that aim to block harmful usage by assuming new harmful behavior arises only through fine‑tuning. It demonstrates that simple, low‑cost attacks—abliteration and prefilling—can significantly increase attack success rates across multiple benchmarks, and proposes an abliteration‑resistant tuning (ART) method that mitigates these attacks. The study shows that safeguarding defenses must be evaluated against a broader range of attack strategies beyond fine‑tuning.
By Kevin Kuo, Virginia Smith, Chhavi Yadav
arXiv:2605.14605v3 Announce Type: replace-cross
Abstract: Model providers increasingly release the weights of large language models. Although these models are safety-aligned before release, their saf...
By Itay Zloczower, Eyal Lenga, Gilad Gressel, Yisroel Mirsky
The paper investigates how inference optimization for large language models can introduce numerical inconsistencies that trigger hidden backdoors. It introduces two types of optimization‑triggered backdoors: the Input‑Specific Optimization Backdoor (ISOB) and the Universal Optimization Backdoor (UOB), the latter enabling a model to remain benign under normal execution but activate a backdoor when optimization is applied. Experiments on seven open‑source LLMs, across multiple tasks and optimization backends, show UOB can achieve up to 100% attack success while maintaining clean accuracy, and the authors propose three defenses that reduce the attack success rate to 0.02.
By Yifei Wang, Yida Yang, Tianlin Li, Xiaohan Zhang, Xiaoyu Zhang, Li Pan
arXiv:2606. 07970v1 Announce Type: cross Abstract: Current open-weight large language models (LLMs) are prone to malicious finetuning attacks, which could compromise the safety alignment of LLMs with only a few steps of supervised finetuning (SFT) on poisoned datasets.
By Haoming Wen, Shi Chen, Qingyu Shi, Siyuan Liu, Minrui Luo, Jingzhao Zhang, Tianxing He
arXiv:2609.05794v1 Announce Type: cross
Abstract: Open-weight large language models face a low-cost white-box threat from representation engineering attacks. Attackers can estimate refusal directions...
By Tian Gao, Zhipeng Xie, Yuhao Wu, Junhua Liu, Xin Fang
The paper introduces DYNASHIELD, a black-box moving target defense for large language models that dynamically customizes decoding hyperparameters and system prompts during inference. It first identifies decoding configurations that lower the probability of successful jailbreak attacks, then samples from a weighted pool of these configurations to add controlled variability to the model’s behavior. Experiments on seven open-source LLMs and four state-of-the-art jailbreak attacks show that DYNASHIELD significantly reduces attack success rates while preserving response quality and adding minimal inference overhead.
By Xiaoqun Liu, Weiming Qi, Qiben Yan