Publications

Xiaoyi Huang, Kejia Zhang, Zhiming Luo (2026). CHASD: Language Increment-Calibrated Contrastive Decoding against Hallucination in LVLMs. Preprint.

Kele Shao, Keda Tao, Kejia Zhang, Sicheng Feng, Mu Cai, Yuzhang Shang, Haoxuan You, Can Qin, Yang Sui, Huan Wang (2026). A Survey of Token Compression for Efficient Multimodal Large Language Models. TMLR.

PDF Cite Github Huggingface OpenReview

Kejia Zhang, Keda Tao, Jiasheng Tang, Huan Wang (2025). Poison as Cure: Visual Noise for Mitigating Object Hallucinations in LVMs. NeurIPS.

PDF Cite Project Github

Kejia Zhang, Keda Tao, Zhiming Luo, Chang Liu, Jiasheng Tang, Huan Wang (2025). TARS: MinMax Token-Adaptive Preference Strategy for MLLM Hallucination Reduction. Preprint.

PDF Cite Project Github

Kejia Zhang, Juanjuan Weng, Shaozi Li, Zhiming Luo (2025). Towards Adversarial Robustness via Debiased High-Confidence Logit Alignment. ICCV.

PDF Cite Github Poster Video

Kejia Zhang, Juanjuan Weng, Yuanzheng Cai, Zhiming Luo, Shaozi Li (2025). Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness. Neural Networks (JCR-Q1).

PDF Cite Github

Kejia Zhang, Junwei Wu, Juanjuan Weng, Guoqing Yang, Zhiming Luo, Shaozi Li (2024). Harmonizing Feature Maps: A Graph Convolutional Approach for Enhancing Adversarial Robustness. Preprint.

PDF Cite Github

Kejia Zhang, Xiao Ding, Bingbing Xiang, Haifeng Zhang, Zhongkui Bao (2024). Extracting Higher Order Topological Semantic via Motif-Based Deep Graph Neural Networks. IEEE Transactions on Computational Social Systems (JCR-Q1).

PDF Cite Github