Publications

Xiaojiang Peng, Jingyi Chen, Zebang Cheng, Bao Peng, Fengyi Wu, Yifei Dong, Shuyuan Tu, Qiyu Hu, Huiting Huang, Yuxiang Lin, Jun-Yan He, Kai Wang, Zheng Lian, Zhi-Qi Cheng

A new framework and benchmark for advancing multimodal emotion understanding across perception and reasoning tasks.

Paper

Fan Zhang*, Zebang Cheng*, Chong Deng*, Haoxuan Li, Zheng Lian, Qian Chen, Huadai Liu, Wen Wang, Yi-Fan Zhang, Renrui Zhang, Ziyu Guo, Zhihong Zhu, Hao Wu, Haixin Wang, Yefeng Zheng, Xiaojiang Peng, Xian Wu, Kun Wang, Xiangang Li, Jieping Ye, Pheng-Ann Heng

* Equal contribution.

A unified benchmark for evaluating emotional understanding and causal reasoning in multimodal large language models across diverse audio-visual scenarios.

Paper Project

Zheng Lian, Xiaojiang Peng, Kele Xu, Ziyu Jia, Xinyi Che, Zebang Cheng, Fei Ma, Laizhong Cui, Yazhou Zhang, Xin Liu, Liang Yang, Jia Li, Fan Zhang, Liumeng Xue, Erik Cambria, Guoying Zhao, Björn W. Schuller, Jianhua Tao

A challenge report and research perspective on the transition from discriminative emotion recognition to open-ended generative understanding.

Paper Challenge

He Hu, Tengjin Weng, Zebang Cheng, Yu Wang, Jiachen Luo, Björn W. Schuller, Zheng Lian, Laizhong Cui

A benchmark focused on how emotions evolve over time and whether multimodal models can explain and predict those transitions.

Paper

Zebang Cheng, Shuimu Chen, Boxue Yang, Yuanshen Guan, Jingyi Chen, Zheng Lian, Xiaojiang Peng, Fei Ma, Laizhong Cui, Qi Tian

An on-policy self-distillation approach that uses privileged rationales to improve affective understanding and reasoning.

Paper

Shuimu Chen, Yuteng Chen, Yuanshen Guan, Zebang Cheng, Zeyu Zhang, Shengqian Qin, Bin Xia, Jiaran Li, Wenming Yang, Fei Ma

An evidence-driven reflection framework for detecting and correcting reasoning errors in long-video understanding.

Paper

Shuimu Chen, Jing Jin, Nan Su, Hongbo Xu, Zebang Cheng, Wenming Yang, Fei Ma, Guijin Wang

A self-distillation framework that transfers language-model judgment capabilities to industrial anomaly detection.

Paper

Zheng Lian, Haoyu Chen, Lan Chen, Haiyang Sun, Licai Sun, Yong Ren, Zebang Cheng, Bin Liu, Rui Liu, Xiaojiang Peng, Jiangyan Yi, Jianhua Tao

Oral presentation.

A unified data, model, and evaluation pipeline for generative multimodal emotion understanding, supported by the MER-Caption dataset and MER-UniBench.

Paper Code

Yuxiang Lin, Jingdong Sun, Zhi-Qi Cheng, Jue Wang, Haomin Liang, Zebang Cheng, Yifei Dong, Jun-Yan He, Xiaojiang Peng, Xian-Sheng Hua

An exploration of emotional reasoning in multimodal large language models beyond conventional emotion classification.

Paper

Yifei Dong, Fengyi Wu, Qi He, Heng Li, Minghan Li, Zebang Cheng, Yuxuan Zhou, Jingdong Sun, Qi Dai, Zhi-Qi Cheng, Alexander G. Hauptmann

A benchmark for embodied agents that must navigate safely and socially around dynamic human participants.

Paper

Fan Zhang, Changhu Wang, Zebang Cheng, Xiaojiang Peng, Dongjie Wang, Yijia Xiao, Chong Chen, Xian-Sheng Hua, Xiao Luo

A semi-supervised framework for aligning 2D images and 3D shapes through decoupled discriminative learning.

Paper

Zi Liang, Peng Hu, Lian Li, Zebang Cheng, Yisu Dong, Haibo He, Qiang Lu, Nan Lin

A vision-language approach for representing EEG signals and supporting automated seizure detection.

Paper

Zheng Lian, Rui Liu, Kele Xu, Bin Liu, Xuefei Liu, Yazhou Zhang, Xin Liu, Yong Li, Zebang Cheng, Haolin Zuo, Ziyang Ma, Xiaojiang Peng, Xie Chen, Ya Li, Erik Cambria, Guoying Zhao, Björn W. Schuller, Jianhua Tao

The MER 2025 challenge report connects established affective-computing tasks with the capabilities of large language models.

Paper Challenge

Dawei Huang, Qing Li, Chuan Yan, Zebang Cheng, Yurong Huang, Xiang Li, Bin Li, Xiaohui Wang, Zheng Lian, Xiaojiang Peng

A hybrid-expert model designed to preserve general visual-language ability while strengthening fine-grained emotional reasoning.

Paper

Zheng Lian, Licai Sun, Haoyu Chen, Zebang Cheng, Fan Zhang, Ziyu Jia, Ziyang Ma, Fei Ma, Xiaojiang Peng, Jianhua Tao

A ranking-based evaluation approach for free-form emotion descriptions when a single definitive ground truth is unavailable.

Paper

Zebang Cheng*, Zhi-Qi Cheng*, Jun-Yan He, Jingdong Sun, Kai Wang, Yuxiang Lin, Zheng Lian, Xiaojiang Peng, Alexander Hauptmann

* Equal contribution.

An early audio-visual large language model for recognizing emotions and producing evidence-grounded reasoning through instruction tuning.

Paper Project Code

Zebang Cheng*, Fuqiang Niu*, Yuxiang Lin, Zhi-Qi Cheng, Bowen Zhang, Xiaojiang Peng

* Equal contribution. 3rd place in the Multimodal Emotion-Cause Analysis Challenge.

A multimodal-language-model system for identifying emotions and their causal utterances in multi-party conversations.

Paper Code

Zebang Cheng*, Shuyuan Tu*, Dawei Huang*, Minghan Li, Xiaojiang Peng, Zhi-Qi Cheng, Alexander G. Hauptmann

* Equal contribution. 1st place in MER-NOISE and 3rd place in the Open-Vocabulary Multimodal Emotion Recognition Challenge.

A Conv-Attention extension of Emotion-LLaMA for robust multimodal emotion recognition under noise and open-vocabulary settings.

Paper Code

Fuqiang Niu, Zebang Cheng, Xianghua Fu, Xiaojiang Peng, Genan Dai, Yin Chen, Hu Huang, Bowen Zhang

A challenge dataset and model for recognizing stance in multimodal, multi-turn conversations.

Paper

Ziheng Qin, Zhaopan Xu, Yukun Zhou, Zangwei Zheng, Zebang Cheng, Hao Tang, Lei Shang, Baigui Sun, Xiaojiang Peng, Radu Timofte, Hongxun Yao, Kai Wang, Yang You

A study of strategies for growing training datasets and improving model performance with scalable data construction.

Paper

Zebang Cheng, Yuxiang Lin, Zhaoru Chen, Xiang Li, Shuyi Mao, Fan Zhang, Daijun Ding, Bowen Zhang, Xiaojiang Peng

2nd place in the ACM Multimedia 2023 Multimodal Emotion Recognition Challenge.

A semi-supervised framework that uses masked autoencoding to learn expressive multimodal representations from limited annotations.

Paper