Junfei Wu

Institute of Automation, Chinese Academy of Sciences

Beijing, China

junfei.wu@cripac.ia.ac.cn

I’m Junfei Wu, a Ph.D. student at the Institute of Automation, Chinese Academy of Sciences (State Key Laboratory of Multimodal Artificial Intelligence), advised by Prof. Tieniu Tan and collaborating with Prof. Qiang Liu, Shu Wu, and Liang Wang. My research focuses on multimodal reasoning in Large Vision-Language Models (LVLMs), including spatial reasoning, hallucination mitigation.

I am currently interning at Qwen-VL, focusing on spatial intelligence. Previously, I interned at Ant NLP Research Group, working on visual reasoning and hallucination in vision-language models.

Outside of research, I enjoy basketball, badminton, table tennis, yo-yo, and Rubik’s Cube.

news

Sep 19, 2025	Our paper “Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing” has been accepted to the NeurIPS 2025. See you in San Diego!
Aug 21, 2025	Our paper “SHARP: Steering Hallucination in LVLMs via Representation Engineering” has been accepted to EMNLP 2025. See you in Suzhou!

selected publications

NeurIPS

Reinforcing spatial reasoning in vision-language models with interwoven thinking and visual drawing

Junfei Wu, Jian Guan, Kaituo Feng, and 5 more authors

In Advances in Neural Information Processing Systems 39 (NeurIPS 2025), 2025

Bib HTML

@inproceedings{wu2025reinforcing,
  title = {Reinforcing spatial reasoning in vision-language models with interwoven thinking and visual drawing},
  author = {Wu, Junfei and Guan, Jian and Feng, Kaituo and Liu, Qiang and Wu, Shu and Wang, Liang and Wu, Wei and Tan, Tieniu},
  booktitle = {Advances in Neural Information Processing Systems 39 (NeurIPS 2025)},
  year = {2025},
}

SHARP: Steering Hallucination in LVLMs via Representation Engineering

Junfei Wu, Yue Ding, Guofan Liu, and 7 more authors

In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025

Bib

@inproceedings{wu2025sharp,
  title = {SHARP: Steering Hallucination in LVLMs via Representation Engineering},
  author = {Wu, Junfei and Ding, Yue and Liu, Guofan and Xia, Tianze and Huang, Ziyue and Sui, Dianbo and Liu, Qiang and Wu, Shu and Wang, Liang and Tan, Tieniu},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},
  pages = {14357--14372},
  year = {2025},
}

Logical Closed Loop: Uncovering Object Hallucinations in Large Vision-Language Models

Junfei Wu, Qiang Liu, Ding Wang, and 4 more authors

In Findings of the Association for Computational Linguistics ACL 2024, 2024

Bib

@inproceedings{wu2024logical,
  title = {Logical Closed Loop: Uncovering Object Hallucinations in Large Vision-Language Models},
  author = {Wu, Junfei and Liu, Qiang and Wang, Ding and Zhang, Jinghao and Wu, Shu and Wang, Liang and Tan, Tieniu},
  booktitle = {Findings of the Association for Computational Linguistics ACL 2024},
  pages = {6944--6962},
  year = {2024},
}

Adversarial contrastive learning for evidence-aware fake news detection with graph neural networks

Junfei Wu, Weizhi Xu, Qiang Liu, and 2 more authors

IEEE Transactions on Knowledge and Data Engineering, 2023

Bib

@article{wu2023adversarial,
  title = {Adversarial contrastive learning for evidence-aware fake news detection with graph neural networks},
  author = {Wu, Junfei and Xu, Weizhi and Liu, Qiang and Wu, Shu and Wang, Liang},
  journal = {IEEE Transactions on Knowledge and Data Engineering},
  volume = {36},
  number = {11},
  pages = {5591--5604},
  year = {2023},
  publisher = {IEEE},
}

Bias mitigation for evidence-aware fake news detection by causal intervention

Junfei Wu, Qiang Liu, Weizhi Xu, and 1 more author

In Proceedings of the 45th International ACM SIGIR conference on research and development in information retrieval, 2022

Bib

@inproceedings{wu2022bias,
  title = {Bias mitigation for evidence-aware fake news detection by causal intervention},
  author = {Wu, Junfei and Liu, Qiang and Xu, Weizhi and Wu, Shu},
  booktitle = {Proceedings of the 45th International ACM SIGIR conference on research and development in information retrieval},
  pages = {2308--2313},
  year = {2022},
}