Publications

Conference Papers

FuseUNet: A Multi-Scale Feature Fusion Method for U-like Networks
Quansong He, Xiangde Min, Kaishen Wang, Tao He
The Forty-Second International Conference on Machine Learning (ICML-2025)
2025
[PDF]

DAMO: Decoding by Accumulating Activations Momentum for Mitigating Hallucinations in Vision-Language Models
Kaishen Wang, Hengrui Gu, Meijun Gao, Kaixiong Zhou
The Thirteenth International Conference on Learning Representations (ICLR-2025)
2025
[PDF]

Strengthening Layer Interaction via Dynamic Layer Attention
Kaishen Wang, Xun Xia, Jian Liu, Zhang Yi, Tao He
Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence (IJCAI-2024)
2024
[PDF]

Preprints

HalluRNN: Mitigating Hallucinations via Recurrent Cross-Layer Reasoning in Large Vision-Language Models
Le Yu^*, Kaishen Wang^*, Jianlong Xiong, Yue Cao, Tao He
arXiv preprint arXiv:2506.17587
2025
[PDF]

Zer0-Jack: A Memory-efficient Gradient-based Jailbreaking Method for Black-box Multi-modal Large Language Models
Tiejin Chen^*, Kaishen Wang^*, Hua Wei
arXiv preprint arXiv:2411.07559
2024
[PDF]