11 papers
Safety in Batches? Understanding and Mitigating Safety Failures in Batch Prompting
Kihyun Kim, Hee-Seon Kim, Wonjun Lee +1
Batch prompting is a practical inference strategy for large language models, but its safety implications remain underexplored. We show that the success of batch prompting for utili…
Exploiting Vision Encoder Vulnerabilities for Universal Adversarial Perturbations on Large Vision-Language Models
Hee-Seon Kim, Minbeom Kim, Seokil Ham +1
Large Vision-Language Models (LVLMs) have achieved remarkable performance on multimodal tasks but remain highly vulnerable to small adversarial perturbations in input images. Exist…
Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-Tuning in Large Language Models
Seokil Ham, Jaehyuk Jang, Wonjun Lee +1
Fine-tuning-as-a-Service (FaaS) enables personalization of large language models (LLMs), but it can weaken safety-alignment under harmful fine-tuning attacks. Recent work has shown…
Safety-Aligned Weights Are Not Enough: Refusal-Teacher-Guided Finetuning Enhances Safety and Downstream Performance under Harmful Finetuning Attacks
Seokil Ham, Yubin Choi, Yujin Yang +3
While Finetuning-as-a-Service (FaaS) enables customization of Large Language Models (LLMs) using user data, this service is vulnerable to safety degradation when user data includes…
Long-tailed Adversarial Training with Self-Distillation
Seungju Cho, Hongsin Lee, Changick Kim
Adversarial training significantly enhances adversarial robustness, yet superior performance is predominantly achieved on balanced datasets. Addressing adversarial robustness in th…
SELFI: Selective Fusion of Identity for Generalizable Deepfake Detection
Younghun Kim, Minsuk Jang, Myung-Joon Kwon +2
Face identity provides a powerful signal for deepfake detection. Prior studies show that even when not explicitly modeled, classifiers often learn identity features implicitly. Thi…