Research

Publications

Selected research on trustworthy multimodal models, adversarial robustness, efficient AI, and graph learning.

Collaboration of LMs and SMs for domain tasks

Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks

Yang Liu^†, Kejia Zhang^†, Bingjie Yan, Tianyuan Zou, Jianqing Zhang, Zixuan Gu, Xiangsen Chen, Jianbing Ding, Xidong Wang, Jingyi Li, Xiaozhou Ye, Ye Ouyang, Qiang Yang, Ya-Qin Zhang († Equal contribution)

arXiv 2026 (Preprint)

Teaser of TARS

TARS: MinMax Token-Adaptive Preference Strategy for MLLM Hallucination Reduction

Kejia Zhang, Keda Tao, Zhiming Luo, Chang Liu, Jiasheng Tang, Huan Wang

European Conference on Computer Vision 2026 (ECCV 2026)

PDF Project GitHub

Overview of CHASD

CHASD: Language Increment-Calibrated Contrastive Decoding against Hallucination in LVLMs

Xiaoyi Huang, Kejia Zhang, Zhiming Luo

arXiv 2026 (Preprint)

A Survey of Token Compression for Efficient Multimodal Large Language Models

A Survey of Token Compression for Efficient Multimodal Large Language Models

Kele Shao, Keda Tao, Kejia Zhang, Sicheng Feng, Mu Cai, Yuzhang Shang, Haoxuan You, Can Qin, Yang Sui, Huan Wang

Transactions on Machine Learning Research 2026 (TMLR 2026)

PDF GitHub Hugging Face OpenReview

Overview of MANI-Pure

MANI-Pure: Magnitude-Adaptive Noise Injection for Adversarial Purification

Xiaoyi Huang, Junwei Wu, Kejia Zhang, Carl Yang, Zhiming Luo

arXiv 2025 (Preprint)

Teaser of VAP

Poison as Cure: Visual Noise for Mitigating Object Hallucinations in LVMs

Kejia Zhang, Keda Tao, Jiasheng Tang, Huan Wang

Advances in Neural Information Processing Systems 2025 (NeurIPS 2025)

PDF Project Poster GitHub

High-confidence logits alignment

Towards Adversarial Robustness via Debiased High-Confidence Logit Alignment

Kejia Zhang, Juanjuan Weng, Shaozi Li, Zhiming Luo

IEEE/CVF International Conference on Computer Vision 2025 (ICCV 2025)

PDF GitHub Poster Video

Model Overview

Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness

Kejia Zhang, Juanjuan Weng, Yuanzheng Cai, Zhiming Luo, Shaozi Li

Neural Networks 2025 (NN 2025)

Model Overview

Extracting Higher Order Topological Semantic via Motif-Based Deep Graph Neural Networks

Kejia Zhang, Xiao Ding, Bingbing Xiang, Haifeng Zhang, Zhongkui Bao

IEEE Transactions on Computational Social Systems 2024 (IEEE TCSS 2024)