Xintong Wang
| Name: | Xintong Wang | ![]() |
| Position: | Ph.D. Candidate | |
| Email: | xintong.wang@uni-hamburg.de | |
| Phone: | +49 40 42883-2322 | |
| Office: | F435 | |
| Address: |
Informatikum, Vogt-Kölln-Straße 30, 22527 Hamburg |
I am a final-year Ph.D. candidate at the University of Hamburg (Exzellenzuniversitäten), Language Technology Group, advised by Prof. Dr. Chris Biemann, who is the best supervisor I've ever had. Before I join LT group, I got the M.Eng. and B.Eng. in Computer Science at the South China University of Technology (Double First-Class Universities and Disciplines) in 2019 and 2016 respectively. I have worked with Prof. Wenjie Li (She taught me how to do research and be critical.) at the Hong Kong Polytechnic University and Prof. Min Yang (She introduced NLP to me.) at the Chinese Academy of Sciences.
Research Interests:
My research focuses on large language models and agentic systems, with an emphasis on training and alignment, evaluation and interpretability, and multilingual and multimodal learning for real-world applications.
Opening!!! We are hiring student assistants at the University of Hamburg. Self-motivated students with experience in LLMs, LVLMs, and Agents are welcome to reach out. Please feel free to drop me an email. I’m always open to collaborations.
News:
- Aug. 2025: One paper accepted to EMNLP 2025 (Main & Top 15%)!
- May 2025: Three papers accepted to ACL Findings 2025!
- Dec. 2024: I will serve as an area chair for ACL ARR!
Selected Publications during PhD: [Full Publication List]
- Xintong Wang, Jingheng Pan, Longqin Jiang, Liang Ding, Xingshan Li, and Chris Biemann, CogSteer: Cognition-Inspired Selective Layer Intervention for Efficient Semantic Steering in Large Language Models, Findings of the Association for Computational Linguistic 2025. (ACL 2025). (CORE Rank A*). [PDF]
- Xintong Wang, Yixiao Liu, Jingheng Pan, Liang Ding, Longyue Wang, and Chris Biemann, Mitigating Biases in Chinese Toxic Language via Sentiment Polarity Consistent Rewrites, The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025). (CORE Rank A*) [PDF].
- Xintong Wang, Jingheng Pan, Yixiao Liu, Xiaohu Zhao, Chenyang Lyu, Minghao Wu, Chris Biemann, Longyue Wang, Linlong Xu, Weihua Luo, Kaifu Zhang, Rethinking Multilingual Vision-Language Translation: Dataset, Evaluation, and Adaptation. [PDF]
- Xintong Wang, Jingheng Pan, Liang Ding, and Chris Biemann, Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding, Findings of the Association for Computational Linguistic 2024. (ACL 2024). (CORE Rank A*)
- Xintong Wang, Xiaoyu Li, Xingshan Li, and Chris Biemann. Probing Language Models from A Human Behavioral Perspective. Proceedings of Bridging Neurons and Symbols for Natural Language Processing and Knowledge Graphs Reasoning (Neusymbridge) at the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024).
- Xintong Wang, Xiaoyu Li, Liang Ding, Sanyuan Zhao, and Chris Biemann. Using Self-Supervised Dual Constraint Contrastive Learning for Cross-modal Retrieval. European Conference on Artificial Intelligence 2023. (ECAI 2023). (CORE Rank A - 24%)
- Xintong Wang*, Florian Schneider*, Özge Alaçam, Prateek Chaudhury, and Chris Biemann. MOTIF: Contextualized Images for Complex Words to Improve Human Reading. Language Resources and Evaluation Conference 2022. (LREC 2022). (CORE Rank B) [MOTIF Dataset]
Professional Activities:
- Invited Talk: Probing Large Language Models (LLMs) for Predicting Human Behavioral Data, June 2022, University of Wuppertal
- Area Chair: NAACL 25'
- Conference Reviewer: ACL 25', EMNLP 25', NAACL 25', COLING 25', AISTATS 25', EMNLP 24', NeurIPS 24', ECCV 24', ACM MM 24', ACL 24', LREC-COLING 24', CVPR24', AACL23', EMNLP 23', ECAI 23', ACM MM23', ACL23‘, EACL 22', AACL 22', EMNLP 22', EMNLP 21', ACL 21', NAACL 21', IJCAI 21' (Senior PC), ACL 20', EMNLP 20', AACL-IJCNLP 20', EMNLP-IJCNLP 19', NLPCC 19'
- Journal Reviewer: ACL Rolling Review, ACM Transactions on Asian and Low-Resource Language Information Processing (TALLIP), IEEE Access
