徐宁
职称：副高
所在院系：计算机科学系
研究方向：机器学习、数据挖掘、大模型
电话：
邮箱：xning@seu.edu.cn
职务：
个人简介

徐宁，bw必威西汉姆联官网计算机科学与工程学院副教授、博导，国家EDA技术创新中心 PI，bw必威西汉姆联官网新一代人工智能技术与交叉应用教育部重点实验室主任助理，bw必威西汉姆联官网“紫金青年学者”。长期从事人工智能领域的研究，在 ICML、NeurIPS、ICLR、IEEE TPAMI 等人工智能领域国际顶级会议和期刊上发表论文 60 余篇，带领团队研发了芯片设计领域大模型 ChipExpert。获 2021 年 CCF 优秀博士学位论文奖，担任德国学术交流中心 AInet Fellow，机器学习三大会 ICML、NeurIPS、ICLR 领域主席、国际期刊 Electronics 编委、中国图象图形学学会（CSIG）国际合作与交流工委会秘书长、中国计算机学会（CCF）人工智能与模式识别专委会执行委员，指导学生获得基金委青年学生基础研究项目、中国科协青年人才托举工程博士生专项等，更多信息请查看个人主页。


欢迎对机器学习、大模型、AI4EDA感兴趣的优秀学生加入（联系邮箱：xning AT seu.edu.cn），组内提供充足的大模型算力资源和充分的科研指导，可做出科研成果并发表于顶会顶刊，详见：学生培养 

  • 招收2026级博士生，还剩1个名额，参加2026年春季博士补招

  • 招收2027级直博生、硕士生（计算机学院、软件学院、东大-蒙纳士联合生院）

  •  每年招收3名本科生进行科研训练，欢迎有兴趣的同学报名（请附上简历、本科成绩及排名）

研究方向
我的研究方向包括机器学习、大语言模型（LLM）以及AI4EDA：

1. 机器学习：主要围绕歧义性与不确定性问题展开研究，以提升模型鲁棒性，提出了面向复杂语义的新型学习方法 —— 标记增强（Label Enhancement）学习。


2. 大语言模型：研究重点为安全可控的大模型系统及数据生成与合成技术。


3. AI for EDA（人工智能辅助电子设计自动化）：致力于利用大语言模型实现芯片设计流程的创新与自动化，团队研发并开源了ChipExpert—— 全球首个面向芯片设计领域的开源大模型。

教育经历
工作经历
科研项目
论文著作

Selected Publication:
(* indicates the corresponding author)

  • J. Yang, N. Xu*, B. Liu, S. Qiao, X. Geng. Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization. In: Proceedings of the 14th International Conference on Learning Representations (ICLR 2026), Brazil, 2026, in press.

  • B. Liu, N. Xu*, J. Wang, X. Geng. Can Class-Priors Help Single-Positive Multi-Label Learning? In: Advances in Neural Information Processing Systems 38 (NeurIPS 2025), San Diego, USA, 2025, in press. (CCF-A)

  • C. Qiao, N. Xu*, Y. Hu, X. Geng. Reduction-based Pseudo-label Generation for Instance-dependent Partial Label Learning. In: Advances in Neural Information Processing Systems 38 (NeurIPS 2025), San Diego, USA, 2025, in press. (CCF-A)

  • T. Wu, S. Zhu, J. Wang, N. Xu*, G. Qi, H. Wang. Uncertain Knowledge Graph Completion via Semi-Supervised Confidence Distribution Learning. In: Advances in Neural Information Processing Systems 38 (NeurIPS 2025), San Diego, USA, 2025, in press. (CCF-A, Spotlight)

  • B. Liu, N. Xu*, X. Geng. Progressively Label Enhancement for Large Language Model Alignment. In: Proceedings of the 42nd International Conference on Machine Learning (ICML 2025), Vancouver, Canada, 2025, in press. (CCF-A)

  • Y. Hu, C. Qiao, X. Geng, N. Xu*. Selective Label Enhancement Learning for Test-Time Adaptation. In: Proceedings of the 13th International Conference on Learning Representations (ICLR 2025), Singapore, 2025.

  • J. Wang, N. Xu*, X. Geng. VADIS: Investigating Inter-View Representation Biases for Multi-View Partial Multi-Label Learning. In: Proceedings of the 41st Conference on Uncertainty in Artificial Intelligence (UAI 2025), Rio de Janeiro, Brazil, 2025, in press. (CCF-B)

  • N. Xu, C. Qiao, Y. Zhao, X. Geng, M.-L. Zhang. Variational Label Enhancement for Instance-Dependent Partial Label Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2024, 46(12): 11298-11313. (CCF-A)

  • J. Lv, B. Liu, L. Feng, N. Xu, M. Xu, B. An, G. Niu, X. Geng, M. Sugiyama. On the Robustness of Average Losses for Partial-Label Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2024, 46(5): 2569-2583. (CCF-A)

  • Y. Liu, J. Lv, X. Geng, N. Xu*. Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency. In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria, 2024, 31614-31628. (CCF-A)

  • B. Liu, N. Xu*, X. Fang, X. Geng. Correlation-induced label prior for semi-supervised multi-label learning. In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria, 2024. (CCF-A)

  • C. Qiao, N. Xu*, Y. Hu, X. Geng. ULAREF: A unified label refinement framework for learning with inaccurate supervision. In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria, 2024. (CCF-A)

  • N. Xu, Y. Hu, C. Qiao, X. Geng. Aligned Objective for Soft-Pseudo-Label Generation in Supervised Learning. In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria, 2024, 55033-55047. (CCF-A)

  • J. Lv, Y. Liu, S. Xia, N. Xu, et al. What Makes Partial-Label Learning Algorithms Effective? In: Advances in Neural Information Processing Systems 37 (NeurIPS 2024), Vancouver, Canada, 2024. (CCF-A)

  • B. Liu, N. Xu*, J. Lv, X. Geng. Revisiting Pseudo-Label for Single-Positive Multi-Label Learning. In: Proceedings of the 40th International Conference on Machine Learning (ICML 2023), Honolulu, Hawaii, USA, 2023, 22249-22265. (CCF-A)

  • C. Qiao, N. Xu*, J. Lv, Y. Ren, X. Geng. FREDIS: A Fusion Framework of Refinement and Disambiguation for Unreliable Partial Label Learning. In: Proceedings of the 40th International Conference on Machine Learning (ICML 2023), Honolulu, Hawaii, USA, 2023, 28321-28336. (CCF-A)

  • N. Xu, B. Liu, J. Lv, C. Qiao, X. Geng. Progressive Purification for Instance-Dependent Partial Label Learning. In: Proceedings of the 40th International Conference on Machine Learning (ICML 2023), Honolulu, Hawaii, USA, 2023, 38551-38565. (CCF-A)

  • C. Qiao, N. Xu*, X. Geng. Decompositional Generation Process for Instance-Dependent Partial Label Learning. In: Proceedings of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda, 2023. (Spotlight)

  • N. Xu, J. Shu, R.-Y. Zheng, X. Geng, D. Meng, M.-L. Zhang. Variational Label Enhancement. IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2023, 45(5): 6537-6551. (CCF-A)

  • H. Yuan, Y. Shi, N. Xu, X. Yang, X. Geng, Y. Rui. Learning from biased soft labels. In: Advances in Neural Information Processing Systems 36 (NeurIPS 2023), New Orleans, USA, 2023, 59566-59584. (CCF-A)

  • N. Xu, C. Qiao, J. Lv, X. Geng, M.-L. Zhang. One Positive Label is Sufficient: Single-Positive Multi-Label Learning with Label Enhancement. In: Advances in Neural Information Processing Systems 35 (NeurIPS 2022), New Orleans, USA, 2022, 21765-21776. (CCF-A, Oral)

  • N. Xu, Y.-P. Liu, and X. Geng. Label Enhancement for Label Distribution Learning. IEEE Transactions on Knowledge and Data Engineering (IEEE TKDE), 2021, 33(4): 1632-1643. (CCF-A)

  • N. Xu, C. Qiao, X. Geng, M.-L. Zhang. Instance-Dependent Partial Label Learning. In: Advances in Neural Information Processing Systems 34 (NeurIPS 2021), Virtual, 2021, 27119-27130. (CCF-A, Spotlight)

  • N. Xu, J. Shu, Y.-P. Liu, X. Geng. Variational Label Enhancement. In: Proceedings of the 37th International Conference on Machine Learning (ICML 2020), Vienna, Austria (Virtual), 2020, 10597-10606. (CCF-A)

 详见课题组近期发表的论文

专利
获奖情况

培养学生：

  • Cong-Yu Qiao (乔聪玉)
    2 ICML, 1 NeurIPS, 2 ICLR

  • Biao Liu (刘彪)
    4 ICML

  • Yihao Hu (胡益豪)
    1 ICML, 1 ICLR

  • Jia-Yu Li (李加羽) → Ph.D. at University of Queensland
    1 IEEE TNNLS

  • Yong-Di Wu (吴永迪) → Ph.D. at Huazhong University of Science and Technology
    1 IEEE TKDE

  • Yangfan Liu (刘杨帆) → Alibaba
    1 ICML, 1 NeurIPS

  • Jie Wang (王洁) → Kuaishou
    1 NeurIPS, 1 UAI

  • Yuchen Zhao (赵宇晨)
    1 TPAMI

  • Shiqi Qiao (乔世琦)
    1 ICLR

  • Yifan Wang (王一帆)
    1 FCS

  • Yihao Hu (胡益豪) → Ph.D. at Southeast University
    1 ICML，1ICLR

  • Junming Yang (杨骏铭)
    1 ICLR



  • 联系方式
  • 通信地址：南京市江宁区bw必威西汉姆联官网路2号bw必威西汉姆联官网九龙湖校区计算机学院
  • 邮政编码：211189
  • ​办公地点：bw必威西汉姆联官网九龙湖校区计算机楼
  • 学院微信公众号