Mambair: A simple baseline for image restoration with state-space model

H Guo, J Li, T Dai, Z Ouyang, X Ren, ST Xia - European Conference on …, 2025 - Springer
Recent years have seen significant advancements in image restoration, largely attributed to
the development of modern deep neural networks, such as CNNs and Transformers …

Badclip: Dual-embedding guided backdoor attack on multimodal contrastive learning

S Liang, M Zhu, A Liu, B Wu, X Cao… - Proceedings of the …, 2024 - openaccess.thecvf.com
While existing backdoor attacks have successfully infected multimodal contrastive learning
models such as CLIP they can be easily countered by specialized backdoor defenses for …

Backdoor Attacks and Defenses Targeting Multi-Domain AI Models: A Comprehensive Review

S Zhang, Y Pan, Q Liu, Z Yan, KKR Choo… - ACM Computing …, 2024 - dl.acm.org
Since the emergence of security concerns in artificial intelligence (AI), there has been
significant attention devoted to the examination of backdoor attacks. Attackers can utilize …

Backdoorllm: A comprehensive benchmark for backdoor attacks on large language models

Y Li, H Huang, Y Zhao, X Ma, J Sun - arXiv preprint arXiv:2408.12798, 2024 - arxiv.org
Generative Large Language Models (LLMs) have made significant strides across various
tasks, but they remain vulnerable to backdoor attacks, where specific triggers in the prompt …

Test-time backdoor attacks on multimodal large language models

D Lu, T Pang, C Du, Q Liu, X Yang, M Lin - arXiv preprint arXiv …, 2024 - arxiv.org
Backdoor attacks are commonly executed by contaminating training data, such that a trigger
can activate predetermined harmful effects during the test phase. In this work, we present …

Pointncbw: Towards dataset ownership verification for point clouds via negative clean-label backdoor watermark

C Wei, Y Wang, K Gao, S Shao, Y Li… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Recently, point clouds have been widely used in computer vision, whereas their collection is
time-consuming and expensive. As such, point cloud datasets are the valuable intellectual …

Parameter-efficient and memory-efficient tuning for vision transformer: a disentangled approach

T Zhang, J Bai, Z Lu, D Lian, G Wang, X Wang… - … on Computer Vision, 2025 - Springer
Recent works on parameter-efficient transfer learning (PETL) show the potential to adapt a
pre-trained Vision Transformer to downstream recognition tasks with only a few learnable …

Adversarial backdoor defense in clip

J Kuang, S Liang, J Liang, K Liu, X Cao - arXiv preprint arXiv:2409.15968, 2024 - arxiv.org
Multimodal contrastive pretraining, exemplified by models like CLIP, has been found to be
vulnerable to backdoor attacks. While current backdoor defense methods primarily employ …

Energy-latency manipulation of multi-modal large language models via verbose samples

K Gao, J Gu, Y Bai, ST Xia, P Torr, W Liu… - arXiv preprint arXiv …, 2024 - arxiv.org
Despite the exceptional performance of multi-modal large language models (MLLMs), their
deployment requires substantial computational resources. Once malicious users induce …

Adversarial robustness for visual grounding of multimodal large language models

K Gao, Y Bai, J Bai, Y Yang, ST Xia - arXiv preprint arXiv:2405.09981, 2024 - arxiv.org
Multi-modal Large Language Models (MLLMs) have recently achieved enhanced
performance across various vision-language tasks including visual grounding capabilities …