Xuan Feng

Xuan Feng

😊 About me

I am a Ph.D. student in Cyber Security at Jinan University, supervised by Prof. Tianlong Gu, affiliated with the Engineering Research Center of Trustworthy Artificial Intelligence (Ministry of Education). I am currently a visiting student at Nanyang Technological University (NTU), sponsored by the China Scholarship Council (CSC), and am fortunate to be advised by Prof. Bo An (AAAI Fellow) and Assoc. Prof. Nicolas Vun Chan Hua​.

Research interests: Trustworthy AI, with a focus on AI Safety, Content Moderation, and Fairness. I am also interested in multimodal learning and recommender systems, and have collaborated with other researchers on related topics.

I collaborate with Dr. Shuai Zhao on AI safety and trustworthy large language models, and with Dr. Yanhao Jia on multimodal bias.

My research aims to understand and enhance the capabilities of next-generation general AI systems in real-world applications.

🔥 News

  • 2025.06 One paper accepted by Pattern Recognition (PR 2025). 🎉 Appreciate the efforts of the co-author @Hao Li.
  • 2025.05 One paper accepted by The Forty-second International Conference on Machine Learning (ICML 2025). 🎉

📝 Preprints

Preprint
sym

Xuan Feng, Bo An, Tianlong Gu✉, Liang Chang, Fengrui Hao, Peipeng Yu, Shuai Zhao, C2PO: Diagnosing and Disentangling Bias Shortcuts in LLMs. arXiv preprint arXiv:2512.23430.

[Paper]

📝 Publications

PR 2025
sym

Tianlong Gu, Hao Li, Xuan Feng✉, Yiqin Luo, AMGSN: Adaptive Mask-Guide Supervised Network for Debiased Facial Expression Recognition [J], Pattern Recognition (PR’25), 2025: 112023.

[Paper] [Code]

(Corresponding Author)

ICML 2025
sym

Peipeng Yu, Jianwei Fei, Hui Gao, Xuan Feng, Zhihua Xia, Chip Hong Chang, Unlocking the capabilities of vision-language models for generalizable and explainable deepfake detection. Proceedings of the 42nd International Conference on Machine Learning (ICML’25), PMLR 267:72925-72943.

[Paper] [Code]

AAAI 2025
sym

Xuan Feng, Tianlong Gu, Xiaoli Liu, Liang Chang, Learning from Mistakes: Self-correct Adversarial Training for Chinese Unnatural Text Correction[C], the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI’25). 2025, 39(22): 23887-23895.

[Paper][Code]

TAFFC 2024
sym

Tianlong Gu, Mingfeng Feng, Xuan Feng✉, Xuemin Wang. SCARE: A Novel Framework to Enhance Chinese Harmful Memes Detection[J], in IEEE Transactions on Affective Computing (TAFFC’24), 2024. doi: 10.1109/TAFFC.2024.3481419.

[Paper] [Code]

(Corresponding Author)

RecSys 2024 Spotlight
sym

Qin Liu*, Xuan Feng*, Tianlong Gu, and Xiaoli Liu. FairCRS: Towards User-oriented Fairness in Conversational Recommendation Systems[C], In Proceedings of The 18th ACM Conference on Recommender Systems (RecSys’24), October 14–18, 2024, Bari, Italy. ACM, New York, NY, USA, doi: 10.1145/3640457.3688150.

[Paper] [Code]

(Equal contributions)

TASLP 2024
sym

Xuan Feng, Tianlong Gu, Liang Chang, Xiaoli Liu, PROTECT: Parameter-Efficient Tuning for Few-Shot Robust Chinese Text Correction[J], in IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP’24), doi: 10.1109/TASLP.2024.3414333.

[Paper] [Code]

ICDE 2024 Oral
sym

Jiayang Li*, Xuan Feng*, Tianlong Gu, Liang Chang, Dual-Teacher De-biasing Distillation Framework for Multi-domain Fake News Detection[C], 2024 IEEE 40th International Conference on Data Engineering (ICDE’24), Utrecht, Netherlands, 2024, pp. 3627-3639, doi: 10.1109/ICDE60146.2024.00279.

[Paper] [Code]

(Equal contributions)

TAFFC 2023
sym

Xuan Feng, Tianlong Gu, Xuguang Bao, Long Li. Behavior-Based Ethical Understanding in Chinese Social News[J], in IEEE Transactions on Affective Computing (TAFFC’23), vol. 14, no. 3, pp. 2349-2360, 1 July-Sept. 2023, doi: 10.1109/TAFFC.2022.3160745.

[Paper] [Code]

  • Peipeng Yu, Jiajan Chen, Xuan Feng, Zhihua Xia. CHEAT: A Large-scale Dataset for Detecting CHatGPT-writtEn AbsTracts[J]. IEEE Transactions on Big Data, 2025 (01): 1-9. [Paper] [Code]

  • 古天龙, 冯旋, 李龙, 包旭光, 李云辉. 基于社会新闻数据集的伦理行为判别方法[J]. 计算机研究与发展, 2021, 58(02): 253-263. doi: 10.7544/issn1000-1239.2021.20200727. [Paper]

🎖 Honors and Awards

  • 2021.10 National Scholarship
  • 2024.10 National Scholarship