Multimodal Robust Prompt Distillation for 3D Point Cloud Models

Xiang Gu; Liming Lu; Xu Zheng; Anan Du; Yongbin Zhou; Shuchao Pang

AAAI 2026technical0 citations

Multimodal Robust Prompt Distillation for 3D Point Cloud Models

Xiang Gu, Liming Lu, Xu Zheng, Anan Du, Yongbin Zhou, Shuchao Pang

Abstract

Adversarial attacks pose a significant threat to learning-based 3D point cloud models, critically undermining their reliability in security-sensitive applications. Existing defense methods often suffer from (1) high computational overhead and (2) poor generalization ability across diverse attack types. To bridge these gaps, we propose a novel yet efficient teacher-student framework, namely Multimodal Robust Prompt Distillation (MRPD) for distilling robust 3D point cloud model. It learns lightweight prompts by aligning student point cloud model

BibTeX

@inproceedings{aaai2026_multimodalrobust,
  title = {Multimodal Robust Prompt Distillation for 3D Point Cloud Models},
  author = {Xiang Gu and Liming Lu and Xu Zheng and Anan Du and Yongbin Zhou and Shuchao Pang},
  booktitle = {AAAI 2026},
  year = {2026}
}