5 papers
Learning Normal Diffusion Dynamics for Backdoor Defense in Text-to-Image Models
Junjian Li, Xiaolong Liu, Peng Sun +4
Backdoor attacks pose a serious threat to the secure deployment of text-to-image (T2I) diffusion models. Existing defenses typically detect backdoors from specific abnormal pattern…
Bit-Flip Attacks on Vision-Language-Action Models: Action-Decoding Architecture Shapes the Vulnerability
Yudong Gao, Linghan Chen, Wenhan Wu +5
Quantized Vision-Language-Action (VLA) models expose a weight-fault surface: Rowhammer-style faults can corrupt deployed INT8 bits. We present the first bit-flip attack on a VLA: a…
Shortcut Before Circuit: Document Statistics Time In-Context Conflict Resolution
Yijun Liao, Fanwei Liang
When a context asserts two values for one fact, a model commits to a cue -- recency, repetition, position -- but natural data rarely makes these disagree, so behavior cannot reveal…
FFCBA: Feature-based Full-target Clean-label Backdoor Attacks
Yangxu Yin, Honglong Chen, Yudong Gao +4
Backdoor attacks pose a significant threat to deep neural networks, as backdoored models would misclassify poisoned samples with specific triggers into target classes while maintai…
SFIBA: Spatial-based Full-target Invisible Backdoor Attacks
Yangxu Yin, Honglong Chen, Yudong Gao +3
Multi-target backdoor attacks pose significant security threats to deep neural networks, as they can preset multiple target classes through a single backdoor injection. This allows…