Mouxing Yang

Mouxing Yang

Sichuan University

Hi, I am Mouxing Yang (杨谋星). I am currently a Research Fellow / Postdoctoral Fellow at the XLearning research group, College of Computer Science, Sichuan University. I obtained my Ph.D. degree from Sichuan University in Dec. 2025 (ahead of schedule，提前毕业) under the supervision of Prof. Xi Peng (彭玺). My work has received over 2,300 citations and has been recognized among the World’s Top 2% Scientists and the National Postdoctoral Innovation Talents Support Program, which selects only 18 candidates annually in the Computer Science and Technology discipline across China.

I have authored more than 30 papers in tier-1 journals and conferences, including 8 IEEE TPAMI/IJCV papers, 2 ICLR Oral papers (~ top 1% of accepted papers). My work has received over 2,300 citations and has been awarded the World’s Top 2% of Scientists and the National Postdoctoral Innovation Talents Support Program. My research mainly focuses on Reliable Multi-modal Learning, Multi-modal Test-time Computing/Adaptation and Agentic AI in the Wild, with contribution in:

  • Reliable Multi-modal Learning: I have worked extensively on tackling challenges such as noisy correspondence (NeurIPS’26, ICLR’26, TPAMI’26, TPAMI’25, IJCV’24, TIP’24, NeurIPS’24, ICCV’23, etc.) and missing modality (TPAMI’26, AAAI’24, TPAMI’23, TPAMI’22) in the image-text, multi-view data. For more insights on noisy correspondence, please refer to our repository: Noisy Correspondence Summary.
  • Multi-modal Test-time Computing/Adaptation: I believe the future of foundation models lies in self-evolving systems. My recent research focuses on unleashing their potential during inference by addressing reliability challenges in dynamic environments, with applications in multi-modal recognition (ICLR’24, AAAI’26), cross-modal retrieval (TPAMI’26, ICLR’25, ICML’25), and multi-modal CoT (ICML’26).
  • Agentic AI in the Wild: I am currently exploring how to extend LLM agents to diverse, open-ended real-world scenarios, with recent work on skill retrieval, software engineering, and efficient coding agents. More studies are underway.

杨谋星，四川大学计算机学院研究员（专职科研）、研究生导师，入选国家博新计划、全球前2%顶尖科学家年度榜单，主要研究方向为鲁棒多模态学习（缺失模态、噪声关联）、多模态测试时自适应、Agentic AI。已在Nature子刊及CCF-A类刊物上发表论文30余篇，谷歌学术引用2300余次，获国自然青C项目、国自然博士生项目、第十二届百度奖学金（全球10人，我国中西部地区唯一）等的资助。

招生要求：如您对我研究方向感兴趣，且有意愿与我一起开展研究，请先阅读XLearning实验室招生细则，随后联系 (Email: yangmouxing[at]gmail.com, yangmouxing[at]scu.edu.cn)

🔥 News

  • 2026.09:   Two NeurIPS’26 paper has been accepted. Congrats to Zhichao and Yijie.
  • 2026.08:   I will serve as Area Chair for ICLR’27.
  • 2026.08:   One TPAMI’27 paper has been accepted. Congrats to Haobin.
  • 2026.08:   I will serve as SPC member for AAAI’27.
  • 2026.05:   Two ICML’26 paper have been accepted. Congrats to Haobin and Yuxin.
  • 2026.02:   Two ICLR’26 paper have been accepted as Orals (~1.3%). Big Congrats to Haobin and Haochen.
  • 2025.12:   One TPAMI’26 paper has been accepted. Congrats to Haobin.
  • 2025.10:   One AAAI’26 paper has been accepted. Congrats to Yutong.
  • 2025.05:   One IJCAI’25 paper and two ICML’25 have been accepted. Congrats to Yiding and Guofeng.
  • 2025.02:   I am honored to be awarded the Baidu Research Fellowship (百度奖学金，10 PhD candidates worldwide).
  • 2025.01:   One paper has been accepted by ICLR’25 as Spotlight (acc rate=5.1%). Congrats to Haobin.
  • 2024.09:   One TPAMI’25 paper and one NeurIPS’24 paper have been accepted.
  • 2024.02:   One paper has been accepted by TIP’24.
  • 2024.01:   One ICLR’24 paper and one IJCV’24 paper have been accepted.
  • 2023.12:   One AAAI’24 paper and one TPAMI’24 paper have been accepted. Congrats to Yiding.
  • 2023.09:   One paper has been accepted by Nature Communications. Big congrats to Yunfan.
  • 2023.07:   One paper has been accepted by ICCV’23. Congrats to Yijie.
  • 2023.04:   One paper has been accepted by IJCAI’23. Congrats to Haobin.
  • 2022.06:   One paper has been accepted by IJCV’22. Congrats to Yunfan.
  • 2022.03:   One paper has been accepted by CVPR’22.
  • 2022.02:   One paper has been accepted by TPAMI’23.
  • 2021.03:   One paper has been accepted by CVPR’21.

🎖 Honors and Awards

  • World’s Top 2% Scientists (全球前 2% 顶尖科学家榜单)
  • National Postdoctoral Innovation Talents Support Program (国家博士后创新人才支持计划)
  • 四川大学第二批“海纳博士后”支持计划 (全校9人)
  • Baidu Research Fellowship (百度奖学金，10 PhD candidates worldwide)
  • Fundamental Research Project for Young Ph.D. students from NSFC (国家自然科学基金青年学生基础研究项目(博士生))
  • National Scholarship (国家奖学金， three times)

💬 Invited Talks

  • 2026.07, Imperfect Chain-of-Thoughts in Multi-modal Reasoning, Special Sessions of ChinaMM 2026, Taiyuan
  • 2025.09, Learning with Noisy Correspondence, Special Sessions of the 22nd Annual Academic Conference on Signal Processing of CIE, Beijing
  • 2022.06, Noisy Correspondences: A New Paradigm for Learning with Noisy Labels, TrustML Young Scientist Seminar, RIKEN AIP (Online)
  • 2022.05, CVPR 2022 Pre-conference Sharing Session, Chengdu

🙋 Service

  • Conference AC/SPC: ICLR, AAAI

  • Journal Reviewer: IEEE TPAMI, IEEE TIP, IEEE TKDE, etc.

  • Conference Reviewer: ICLR, NeurIPS (Top reviewer), ICML (Gold reviewer), CVPR, ICCV, etc.