Wei Wang Postdoctoral Researcher Imperfect Information Learning Team Center for Advanced Intelligence Project RIKEN Tokyo 103-0027, Japan Email: wei.wang [at] riken.jp OR wwangwitsel [at] gmail.com

Biography

I am a postdoctoral researcher at the Imperfect Information Learning Team, RIKEN Center for Advanced Intelligence Project (AIP). I received my Ph.D. degree from the University of Tokyo, where I was very fortunate to be supervised by Prof. Masashi Sugiyama.

My research interests lie in machine learning and large language models (LLMs). Currently, I am interested in LLM post-training and LLM reasoning. Previously, my research focused mainly on data-centric machine learning.

Publication

Conference Paper

• Realistic Evaluation of Deep Partial-Label Learning Algorithms.
  W. Wang, D.-D. Wu, J. Wang, G. Niu, M.-L. Zhang, and M. Sugiyama.
  In: Proceedings of the 13th International Conference on Learning Representations (ICLR 2025), Singapore, 2025.
  [Paper] [Code] [Data] [Spotlight]
• Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical.
  W. Wang, T. Ishida, Y.-J. Zhang, G. Niu, and M. Sugiyama.
  In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), pp. 50683-50710, Vienna, Austria, 2024.
  [Paper] [Code]
• Binary Classification with Confidence Difference.
  W. Wang, L. Feng, Y. Jiang, G. Niu, M.-L. Zhang, and M. Sugiyama.
  In: Advances in Neural Information Processing Systems 36 (NeurIPS 2023), pp. 5936-5960, New Orleans, LA, 2023.
  [PDF] [Code]
• Label Specific Multi-Semantics Metric Learning for Multi-Label Classification: Global Consideration Helps.
  J.-X. Mao, W. Wang, and M.-L. Zhang.
  In: Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023), pp. 4055-4063, Macau, China, 2023.
  [Paper] [Code]
• Partial Label Learning with Discrimination Augmentation.
  W. Wang, and M.-L. Zhang.
  In: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022), pp. 1920-1928, Washington, DC, 2022.
  [Paper] [Code]
• Semi-Supervised Partial Label Learning via Confidence-Rated Margin Maximization.
  W. Wang, and M.-L. Zhang.
  In: Advances in Neural Information Processing Systems 33 (NeurIPS 2020), pp. 6982-6993, Online, 2020.
  [Paper] [Code]

Journal Article

• Weakly Supervised Classification with Pre-Trained Models: A Robust Fine-Tuning Approach.
  M. Li*, W. Wang*, and M. Sugiyama. (* indicates equal contribution)
  Machine Learning, in press.
  [Paper] [Code]

Preprint

• Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms.
  W. Wang*, D.-D. Wu*, M. Li, J. Zhang, G. Niu, and M. Sugiyama.
  [arXiv]
• Rethinking Consistent Multi-Label Classification under Inexact Supervision.
  W. Wang*, T. Ma*, M.-K. Xie, G. Niu, and M. Sugiyama.
  [arXiv]

Talk

• On the Relation between Complementary-Label Learning and Negative-Unlabeled Learning.
  2023.11.24, International Workshop on Weakly Supervised Learning 2023.

Award

• Outstanding Reviewer, KDD 2025.
• RIKEN Ohbu Award (Researcher Incentive Award), 2024.
• Scholar/Travel Award, NeurIPS 2023.
• Microsoft Research Asia D-CORE Award, 2023.
• Japanese Government (MEXT) Scholarship, 2023.
• National Scholarship, MOE of PRC, 2021.

Academic Service

• PC Member/Conference Reviewer: ICLR (2024,2025,2026), ICML (2024,2025), NeurIPS (2023,2024,2025), AISTATS (2025,2026), KDD (2023,2024,2025,2026), WWW (2026), SDM (2024), ACML (2024,2025), AAAI (2025,2026).
• Journal Reviewer: IEEE Transactions on Pattern Analysis and Machine Intelligence, Journal of Artificial Intelligence Research, International Journal of Computer Vision, Machine Learning, Neural Networks, SCIENCE CHINA Information Sciences, ACM Transactions on Intelligent Systems and Technology.
• Web Team Member of KDD 2021.