Ph.D. student Knowledge Mining Group Nanjing University of Science and Technology 836178735@qq.com wxy_yyjhl@njust.edu.cn Google Scholar   CV

About me

I am currently a third-year Ph.D. student in Knowledge Mining Group, Nanjing University of Science and Technology.
I am supervised by Prof. Jianfeng Lu and Prof. Yang Yang.
My current research interests are MLLM, Text Generation, Test-Time Adaptation and Prompt Learning.

News

• [08/2025] Four papers was submitted to ICLR 2026.

• [07/2025] Two papers accepted by ACM MM 2025.

• [01/2025] One paper accepted by ICLR 2025.

• [08/2024] One paper accepted by ACML 2024.

• [04/2024] One paper accepted by IJCAI 2024.

• [03/2024] One paper accepted by ICME 2024.

• [06/2022] One paper accepted by ICIP 2022.

Research Intern

• Alibaba International Digital Commerce Group. (2023.08 to Now)
  Working on - Algorithms - Visual & Multimodal.

Publications

• Text as Any-Modality for Zero-Shot Classification by Consistent Prompt Tuning [Github]
  Xiangyu Wu, Feng Yu, Yang Yang*, Jianfeng Lu*
  Proceedings of the 33rd ACM International Conference on Multimedia (ACM MM'2025)




• Noise Self-Correction via Relation Propagation for Robust Cross-Modal Retrieval [Github]
  Ruoxuan Li, Xiangyu Wu, Yang Yang*
  Proceedings of the 33rd ACM International Conference on Multimedia (ACM MM'2025)




• Multi-Label Test-Time Adaptation with Bound Entropy Minimization [Github]
  Xiangyu Wu, Feng Yu, Qing-Guo Chen, Yang Yang*, Jianfeng Lu*
  Proceedings of the 13th International Conference on Learning Representations (ICLR'2025)




• Refining Visual Perception for Decoration Display: A Self-Enhanced Deep Captioning Model [Github]
  Longfei Huang, Xiangyu Wu, Jingyuan Wang, Weili Guo, Yang Yang*
  Proceedings of the 16th Asian Conference on Machine Learning (ACML'2024)




• TAI++: Text as Image for Multi-Label Image Classification by Co-Learning Transferable Prompt [Github]
  Xiangyu Wu, Qingyuan Jiang, Yifeng Wu, Qingguo Chen, Yang Yang*, and Jianfeng Lu*
  Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI'2024)




• CoVLR: Coordinating Cross-Modal Consistency and Intra-Modal Relations for Vision-Language Retrieval [Github]
  Fengqiang Wan, Xiangyu Wu, Zhihao Guan, Yang Yang*
  IEEE International Conference on Multimedia and Expo (ICME'2024)




• QUES-TO-VISUAL GUIDED VISUAL QUESTION ANSWERING
  Xiangyu Wu, Jianfeng Lu, Zhuanfeng Li, Fengchao Xiong
  IEEE International Conference on Image Processing (ICIP'2022)

Awards & Honors

• Excellence Award WWW2025多模态对话系统意图识别挑战赛 2025.01

• First Place 2024第三届 粤港澳大湾区国际算法算例大赛 多模态大模型学科能力综合强化 2024.10

• Second Prize 2024全球人工智能技术创新大赛 赛道1 无人机视角下的双光目标检测 2024.06

• First Place CVPR 2024 New Frontiers for Zero-shot lmage Captioning Evaluation 2024.03

• Second Place 2023第二届 粤港澳大湾区国际算法算例大赛 基于语言增强的图像新类别发现 2024.01

• First Place ICCV 2023 The First Scientific Figure Captioning (SCICAP) Challenge 2023.10

• First Place ICCV 2023 Multi-modal Algorithmic Reasoning (SMART-101) Challenge 2023.10

• First Place 2023全球人工智能技术创新大赛 赛道1 影像学NLP-医学影像诊断报告生成 2023.06

• First Place CVPR 2023 foundation model challenge Cross-Modal Image Retrieval 2023.05

• First Place CVPR 2023 New Frontiers for Zero-shot lmage Captioning Evaluation 2023.05

• Second Place WSDM 2023 Cup: Visual Question Answering Challenge 2023.01

Academic Services

• Journal Reviewer of PR

• Conference Reviewer of AAAI, ICLR, NeurIPS