8 papers
Never Stop Speaking: a Denial-of-Service Attack on End-to-End Speech Language Models
Shuozhe Cheng, Kunlan Xiang, Mingxuan Li +3
Many studies have shown that specially crafted inputs can induce large language models (LLMs) to generate excessively long outputs, resulting in significant computational overhead…
Beyond Native Success: Auditing Deployment-Interface Exposure of CLIP Backdoors
Kunlan Xiang, Haomiao Yang, Wenbo Jiang
Contrastive Language-Image Pre-training models are widely reused across downstream interfaces, including feature extraction, retrieval, reranking, and selection. Existing CLIP back…
DIVER:Diving Deeper into Distilled Data via Expressive Semantic Recovery
Qianxin Xia, Zhiyong Shu, Wenbo Jiang +3
Dataset distillation aims to synthesize a compact proxy dataset that is unreadable or non-raw from the original dataset for privacy protection and highly efficient learning. Howeve…
CBV: Clean-label Backdoor Attacks on Vision Language Models via Diffusion Models
Ji Guo, Xiaolong Qin, Cencen Liu +3
Vision-Language Models (VLMs) have achieved remarkable success in tasks such as image captioning and visual question answering (VQA). However, as their applications become increasi…
UGID: Unified Graph Isomorphism for Debiasing Large Language Models
Zikang Ding, Junchi Yao, Junhao Li +4
Large language models (LLMs) exhibit pronounced social biases. Output-level or data-optimization--based debiasing methods cannot fully resolve these biases, and many prior works ha…
Delayed Backdoor Attacks: Exploring the Temporal Dimension as a New Attack Surface in Pre-Trained Models
Zikang Ding, Haomiao Yang, Meng Hao +6
Backdoor attacks against pre-trained models (PTMs) have traditionally operated under an ``immediacy assumption,'' where malicious behavior manifests instantly upon trigger occurren…