collaborators

8 papers

cs.SD2026

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…

cs.CR2026

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…

cs.CV2026

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…

cs.AI2026

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…

cs.CL2026

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…

cs.CR2026

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…