I serve as an algorithm expert at Alibaba’s Tongyi Laboratory. I received my Ph.D. from Fudan University in 2023, under the supervision of Professor Jianqing Fan and Professor Zhongyu Wei. Before that, I received my Bachelor’s degree in Mathematics and Applied Mathematics from Fudan University. During my Ph.D., I also spent time with MReaL (Machine Reasoning and Learning) at Nanyang Technological University, advised by Hanwang Zhang.

My primary research interests lie in Vision-Language Reasearch including large vision-language model and multi-modal agents. Currently, I focus on the Qwen-VL.

Professional Service

• Area Chair: ACL 2023/2024
• Reviewer: NeurIPS, ICML, ICLR, ACL, EMNLP, IJCAI, AAAI.

Honors and Awards

• Alibaba Star (Top 1%), Alibaba Group, 2023
• ByteDance Scholarship (10 in China), 2021
• National Scholarship for Doctoral Students (Top 1%) in Fudan University, Ministry of Education, 2019

Preprint

• AI Hospital: Interactive Evaluation and Collaboration of LLMs as Intern Doctors for Clinical Diagnosis
  Zhihao Fan, Jialong Tang, Wei Chen, Siyuan Wang, Zhongyu Wei, Jun Xie, Fei Huang, Jingren Zhou.
  Arxiv. [code]

• Benchmark Self-Evolving: A Multi-Agent Framework for Dynamic LLM Evaluation
  Siyuan Wang, Zhuohan Long, Zhihao Fan, Zhongyu Wei.
  Arxiv. [code]

• Reform-Eval: Evaluating Large Vision Language Models via Unified Re-formulation of Task-Oriented Benchmarks
  Zejun Li, Ye Wang, Mengfei Du, Qingwen Liu, Binhao Wu, Jiwen Zhang, Chengxing Zhou, Zhihao Fan, Jie Fu, Jingjing Chen, Xuanjing Huang, Zhongyu Wei.
  Arxiv. [code]

Selected Publications

(Full list see Google Scholar)

• AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation
  Tong Wu*, Zhihao Fan* (Equal Contribution), Xiao Liu, Hai-Tao Zheng, Yeyun Gong, Jian Jiao, Juntao Li, Zhongyu Wei, Jian Guo, Nan Duan, Weizhu Chen.
  NeurIPS 2023. [code]

• Unifying Cross-Lingual and Cross-Modal Modeling Towards Weakly Supervised Multilingual Vision-Language Pre-training
  Zejun Li, Zhihao Fan, Jingjing Chen, Qi Zhang, Xuanjing Huang, Zhongyu Wei.
  ACM MM 2022. [code]

• Negative Sample is Negative in Its Own Way: Tailoring Negative Sentences for Image-Text Retrieval
  Zhihao Fan, Zhongyu Wei, Zejun Li, Siyuan Wang, Jianqing Fan.
  NAACL 2022 Findings. [code]

• MVP: Multi-Stage Vision-Language Pre-training via Multi-Level Semantic Alignment
  Zejun Li, Zhihao Fan, Huaixiao Tou, Zhongyu Wei.
  ACM MM 2022. [code]

• TCIC: Theme Concepts Learning Cross Language and Vision for Image Captioning
  Zhihao Fan, Zhongyu Wei, Siyuan Wang, Ruize Wang, Zejun Li, Haijun Shan, Xuanjing Huang.
  IJCAI 2021. [code]

• Mask Attention Networks: Rethinking and Strengthen Transformer
  Zhihao Fan, Yeyun Gong, Dayiheng Liu, Zhongyu Wei, Siyuan Wang, Jian Jiao, Nan Duan, Xuanjing Huang.
  NAACL 2021. [code]

• Bridging by Word: Image Grounded Vocabulary Construction for Visual Captioning
  Zhihao Fan, Zhongyu Wei, Siyuan Wang, Xuanjing Huang.
  ACL 2019. [code]

• A Reinforcement Learning Framework for Natural Question Generation Using Bi-Discriminators
  Zhihao Fan, Yeyun Gong, Dayiheng Liu, Zhongyu Wei, Siyuan Wang, Jian Jiao, Nan Duan, Xuanjing Huang.
  COLING 2018.

• A Question Type Driven Framework to Diversify Visual Question Generation.
  Zhihao Fan, Zhongyu Wei, Piji Li, Yanyan Lan, Xuanjing Huang.
  IJCAI 2018.

Experience

• Alibaba. Tongyi Lab. Sep 2023 - Present
  • Algorithm Expert.
  • Focus:
    • Large Vision-Language Model
    • Multimodal Agents
• Fudan University. DISC Lab. Oct 2016 - Sep 2023
  • Graduate Research Assistant. Advisor: Zhongyu Wei
  • Focus:
    • Vision-Language Generation and Retrieval
    • Vision-Language Pre-training
• Nanyang Technological University. MReal Lab. Oct 2022 - Apr 2023
  • Visiting Scholar. Advisor: Hangwang Zhang
  • Focus: Diffusion Model
• Microsoft Research Asia. NLC Group. Sep 2019 - Mar. 2020
  • Research Intern. Advisor: Yeyun Gong
  • Focus: Transformer Architecture