3 papers
cs.CR2026
Understanding and Mitigating Prompt Leaking Attacks in Real-World LLM-Based Applications
Yong Yang, Chong Fu, Tong Zhang +6
Large language model (LLM)-based applications rely on system prompts to encode core logic and developer-defined constraints, making these prompts important intellectual property. H…
cs.LG2025
Reformulation is All You Need: Addressing Malicious Text Features in DNNs
Yi Jiang, Oubo Ma, Yong Yang +2
Human language encompasses a wide range of intricate and diverse implicit features, which attackers can exploit to launch adversarial or backdoor attacks, compromising DNN models f…
cs.CV2025
Pre-training CLIP against Data Poisoning with Optimal Transport-based Matching and Alignment
Tong Zhang, Kuofeng Gao, Jiawang Bai +5
Recent studies have shown that Contrastive Language-Image Pre-training (CLIP) models are threatened by targeted data poisoning and backdoor attacks due to massive training image-ca…