Publications

(* Corresponding author; † Equal contribution)

---

Preprints

• Zhifang Zhang, Shuo He*, Haobo Wang, Bingquan Shen, Lei Feng*.
  Defending Multimodal Backdoored Models by Repulsive Visual Prompt Tuning.
  

• Yuwei Niu, Shuo He, Qi Wei, Zongyu Wu, Feng Liu, Lei Feng*.
  BDetCLIP: Multimodal Prompting Contrastive Test-Time Backdoor Detection.
  

• Shuo He, Zhifang Zhang, Feng Liu, Roy Ka-Wei Lee, Bo An, Lei Feng*.
  A Closer Look at Backdoor Attacks on CLIP.
  

• Jiahan Zhang, Shengjie Zhou, Qi Wei, Shuo He, Feng Liu, Lei Feng*.
  Exploring Undetectability and Versatility of Adversarial Examples for Large Multi-Modal Models.
  

• Zongqian Wu, Tianyu Li, Jiaying Yang, Mengmeng Zhan, Xiaofeng Zhu*, Lei Feng*.
  Is Depth All You Need? An Exploration of Iterative Reasoning in LLMs.
  

• Zongqian Wu, Baoduo Xu, Ruochen Cui, Mengmeng Zhan, Xiaofeng Zhu*, Lei Feng*
  Rethinking Chain-of-Thought from the Perspective of Self-Training.
  

---

Selected Publications

Full publications in [Google Scholar Profile]

• Shengjie Zhou, Xin Cheng, Haiyang Xu, Ming Yan, Tao Xiang, Feng Liu, Lei Feng*.
  Endowing Visual Reprogramming with Adversarial Robustness.
  Proceedings of the 13th International Conference on Learning Representations (ICLR'25), 2025.
  

• Suqin Yuan, Runqi Lin, Lei Feng*, Bo Han, Tongliang Liu*.
  Instance-dependent Early Stopping.
  Proceedings of the 13th International Conference on Learning Representations (ICLR'25), 2025.
  

• Chengyi Cai, Zesheng Ye, Lei Feng, Jianzhong Qi, Feng Liu.
  Attribute-based Visual Reprogramming for Image Classification with CLIP.
  Proceedings of the 13th International Conference on Learning Representations (ICLR'25), 2025.
  

• Qi Wei, Shuo He, Jiahan Zhang, Lei Feng*, Bo An.
  Influence-Based Fair Selection for Sample-Discriminative Backdoor Attack.
  Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI'25), accepted, 2025.
  (Oral, Acceptance Rate: 4.63%)

• Haoliang Sun, Qi Wei, Lei Feng, Yupeng Hu, Fan Liu, Hehe Fan, Yilong Yin.
  Variational Rectification Inference for Learning with Noisy Labels.
  International Journal of Computer Vision (IJCV), vol. 133, pp. 652–671, 2025.
  

• Chengyi Cai, Zesheng Ye, Lei Feng, Jianzhong Qi, Feng Liu.
  Bayesian-Guided Label Mapping for Visual Reprogramming.
  Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS'24), accepted, 2024.
  (Oral, Acceptance Rate: 0.41%)

• Zixi Wei, Yuzhou Cao, Lei Feng*.
  Exploiting Human-AI Dependence for Learning to Defer.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 52484-52499, 2024.
  

• Zhenlong Liu, Lei Feng, Huiping Zhuang, Xiaofeng Cao, Hongxin Wei.
  Mitigating Privacy Risk in Membership Inference by Convex-Concave Loss.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 30998-31014, 2024.
  

• Jiahan Zhang, Qi Wei, Feng Liu, Lei Feng*.
  Candidate Pseudolabel Learning: Enhancing Vision-Language Models by Prompt Tuning with Unlabeled Data.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 60004-60020, 2024.
  (Oral, Acceptance Rate: 1.52%)

• Hao Chen, Jindong Wang, Lei Feng, Xiang Li, Yidong Wang, Xing Xie, Masashi Sugiyama, Rita Singh, Bhiksha Raj.
  A General Framework for Learning from Weak Supervision.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 7462-7485, 2024.
  

• Jinhao Li, Haopeng Li, Sarah Monazam Erfani, Lei Feng, James Bailey, Feng Liu.
  Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 28018-28039, 2024.
  

• Changchun Li, Yuanchao Dai, Lei Feng, Ximing Li, Bing Wang, Jihong Ouyang.
  Positive and Unlabeled Learning with Controlled Probability Boundary Fence.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 27641-27652, 2024.
  

• Chengyi Cai, Zesheng Ye, Lei Feng, Jianzhong Qi, Feng Liu.
  Sample-Specific Masks for Visual Reprogramming-based Prompting.
  Proceedings of the 41st International Conference on Machine Learning (ICML'24), pp. 5383-5408, 2024.
  (Spotlight, Acceptance Rate: 3.54%)

• Shiyu Tian, Hongxin Wei, Yiqun Wang, Lei Feng*.
  CroSel: Cross Selection of Confident Pseudo Labels for Partial-Label Learning.
  Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR'24), pp. 19479-19488, 2024.
  (Oral, Acceptance Rate: 0.78%)

• Yuzhou Cao, Lei Feng*, Bo An.
  Consistent Hierarchical Classification with A Generalized Metric.
  Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS'24), pp. 4825-4833, 2024.
  

• Shuqi Liu, Yuzhou Cao, Qiaozhen Zhang, Lei Feng, Bo An.
  Mitigating Underfitting in Learning to Defer with Consistent Losses.
  Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS'24), pp. 4816-4824, 2024.
  

• Suqin Yuan, Lei Feng*, Tongliang Liu*.
  Early Stopping Against Label Noise Without Validation Data.
  Proceedings of the 12th International Conference on Learning Representations (ICLR'24), 2024.
  

• Zixi Wei, Senlin Shu, Yuzhou Cao, Hongxin Wei, Bo An, Lei Feng*.
  Consistent Multi-Class Classification from Multiple Unlabeled Datasets.
  Proceedings of the 12th International Conference on Learning Representations (ICLR'24), 2024.
  (Spotlight, Acceptance Rate: 5.01%)

• Shengjie Zhou, Lue Tao, Yuzhou Cao, Tao Xiang, Bo An, Lei Feng*.
  On the Vulnerability of Adversarially Trained Models Against Two-faced Attacks.
  Proceedings of the 12th International Conference on Learning Representations (ICLR'24), 2024.
  

• Shuo He, Chaojie Wang, Guowu Yang*, Lei Feng*.
  Candidate Label Set Pruning: A Data-centric Perspective for Deep Partial-label Learning.
  Proceedings of the 12th International Conference on Learning Representations (ICLR'24), 2024.
  (Oral, Acceptance Rate: 1.16%)

• Haobo Wang, Ruixuan Xiao, Yixuan Li, Lei Feng, Gang Niu, Gang Chen, Junbo Zhao.
  PiCO+: Contrastive Label Disambiguation for Robust Partial Label Learning.
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 46, no. 5, pp. 3183-3198, 2024.
  

• Jiaqi Lv, Biao Liu, Lei Feng, Ning Xu, Miao Xu, Bo An, Gang Niu, Xin Geng, Masashi Sugiyama.
  On the Robustness of Average Losses for Partial-Label Learning.
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 46, no. 5, pp. 2569-2583, 2024.
  

• Senlin Shu, Deng-Bao Wang, Suqin Yuan, Hongxin Wei, Jiuchuan Jiang, Lei Feng*, Min-Ling Zhang.
  Multiple-Instance Learning from Triplet Comparison Bags.
  ACM Transactions on Knowledge Discovery from Data (TKDD), vol. 18, no. 4, pp. 1-18, 2024.
  

• Beibei Li, Yiyuan Zheng, Beihong Jin, Tao Xiang, Haobo Wang, Lei Feng*.
  AsyCo: An Asymmetric Dual-task Co-training Model for Partial-label Learning.
  Science China Information Sciences (SCIS), accepted, 2024.
  

• Senlin Shu, Haobo Wang, Zhuowei Wang, Bo Han, Tao Xiang, Bo An, Lei Feng*.
  Online Binary Classification from Similar and Dissimilar Data.
  Machine Learning (MLJ), vol. 113, pp. 3463–3484, 2024.
  

• Xin Cheng, Yuzhou Cao, Haobo Wang, Hongxin Wei, Bo An, Lei Feng*.
  Regression with Cost-based Rejection.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 45172-45196, 2023.
  

• Zhiqing Xiao, Haobo Wang, Ying Jin, Lei Feng, Gang Chen, Fei Huang, Junbo Zhao.
  SPA: A Graph Spectral Alignment Perspective for Domain Adaptation.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 37252-37272, 2023.
  

• Yuzhou Cao, Hussein Mozannar, Lei Feng*, Hongxin Wei, Bo An.
  In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defer.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 38485-38503, 2023.
  

• Renchunzi Xie, Hongxin Wei, Lei Feng, Yuzhou Cao, Bo An.
  On the Importance of Feature Separability in Predicting Out-Of-Distribution Error.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 27783-27800, 2023.
  

• Mingyu Xu, Zheng Lian, Lei Feng, Bin Liu, Jianhua Tao.
  ALIM: Adjusting Label Importance Mechanism for Noisy Partial Label Learning.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 38668-38684, 2023.
  

• Wei Wang, Lei Feng, Yuchen Jiang, Gang Niu, Min-Ling Zhang, Masashi Sugiyama.
  Binary Classification with Confidence Difference.
  Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS'23), pp. 5936-5960, 2023.
  

• Suqin Yuan, Lei Feng*, Tongliang Liu*.
  Late Stopping: Avoiding Confidently Learning from Mislabeled Examples.
  Proceedings of the International Conference on Computer Vision (ICCV'23), pp. 16079-16088, 2023.
  

• Hongxin Wei, Huiping Zhuang, Renchunzi Xie, Lei Feng, Gang Niu, Bo An, Yixuan Li.
  Mitigating Memorization of Noisy Labels by Clipping the Model Prediction.
  Proceedings of the 40th International Conference on Machine Learning (ICML'23), pp. 36868-36886, 2023.
  

• Zixi Wei, Lei Feng*, Bo Han, Tongliang Liu, Gang Niu, Xiaofeng Zhu, Heng Tao Shen.
  A Universal Unbiased Method for Classification from Aggregate Observations.
  Proceedings of the 40th International Conference on Machine Learning (ICML'23), pp. 36804-36820, 2023.
  

• Xin Cheng, Yuzhou Cao, Ximing Li, Bo An, Lei Feng*.
  Weakly Supervised Regression with Interval Targets.
  Proceedings of the 40th International Conference on Machine Learning (ICML'23), pp. 5428-5448, 2023.
  

• Shuqi Liu, Yuzhou Cao, Qiaozhen Zhang, Lei Feng, Bo An.
  Consistent Complementary-Label Learning via Order-Preserving Losses.
  Proceedings of the 26th International Conference on Artificial Intelligence and Statistics (AISTATS'23), pp. 8734-8748, 2023.
  

• Xin Cheng, Dengbao Wang, Lei Feng*, Minling Zhang, Bo An.
  Partial-Label Regression.
  Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI'23), pp. 7140-7147, 2023.
  (Acceptance Rate: 19.6%)

• Senlin Shu, Shuo He, Haobo Wang, Hongxin Wei, Tao Xiang, Lei Feng*.
  A Generalized Unbiased Risk Estimator for Learning with Augmented Classes.
  Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI'23), pp. 9829-9836, 2023.
  (Acceptance Rate: 19.6%)

• Lei Feng, Senlin Shu, Yuzhou Cao, Lue Tao, Hongxin Wei, Tao Xiang, Bo An, Gang Niu.
  Multiple-Instance Learning from Unlabeled Bags with Pairwise Similarity.
  IEEE Transactions on Knowledge and Data Engineering (TKDE), vol. 35, no. 11, pp. 11599-11609, 2023.
  

• Hongxin Wei, Renchunzi Xie, Lei Feng*, Bo Han, Bo An.
  Deep Learning from Multiple Noisy Annotators as A Union.
  IEEE Transactions on Neural Networks and Learning Systems (TNNLS), vol. 34, no. 12, pp. 10552-10562, 2023.
  

• Yuzhou Cao, Tianchi Cai, Lei Feng*, Lihong Gu, Jinjie Gu, Bo An, Gang Niu, Masashi Sugiyama.
  Generalizing Consistent Multi-Class Classification with Rejection to be Compatible with Arbitrary Losses.
  Proceedings of the 36th Conference on Neural Information Processing Systems (NeurIPS'22), pp. 521-534, 2022.
  

• Haobo Wang, Mingxuan Xia, Yixuan Li, Yuren Mao, Lei Feng, Gang Chen, Junbo Zhao.
  SoLar: Sinkhorn Label Refinery for Imbalanced Partial-Label Learning.
  Proceedings of the 36th Conference on Neural Information Processing Systems (NeurIPS'22), pp. 8104-8117, 2022.
  

• Lue Tao, Lei Feng, Hongxin Wei, Jinfeng Yi, Shengjun Huang, Songcan Chen.
  Can Adversarial Training Be Manipulated By Non-Robust Features?
  Proceedings of the 36th Conference on Neural Information Processing Systems (NeurIPS'22), pp. 26504-26518, 2022.
  

• Hongxin Wei, Lue Tao, Renchunzi Xie, Lei Feng, Bo An.
  Open-Sampling: Exploring Out-of-Distribution data for Re-balancing Long-tailed datasets.
  Proceedings of the 39th International Conference on Machine Learning (ICML'22), pp. 23615-23630, 2022.
  

• Hongxin Wei, Renchunzi Xie, Hao Cheng, Lei Feng, Bo An, Yixuan Li.
  Mitigating Neural Network Overconfidence with Logit Normalization.
  Proceedings of the 39th International Conference on Machine Learning (ICML'22), pp. 23631-23644, 2022.
  

• Haobo Wang, Ruixuan Xiao, Yixuan Li, Lei Feng, Gang Niu, Gang Chen, Junbo Zhao.
  PiCO: Contrastive Label Disambiguation for Partial-Label Learning.
  Proceedings of the 10th International Conference on Learning Representations (ICLR'22), 2022.
  (Outstanding Paper Award Honorable Mention, Acceptance Rate: <3‰ (10/3391))

• Changchun Li, Ximing Li, Lei Feng, Jihong Ouyang.
  Who Is Your Right Mixup Partner in Positive and Unlabeled Learning?
  Proceedings of the 10th International Conference on Learning Representations (ICLR'22), 2022.
  

• Fei Zhang, Lei Feng*, Bo Han*, Tongliang Liu, Gang Niu, Tao Qin, Masashi Sugiyama.
  Exploiting Class Activation Value for Partial-Label Learning.
  Proceedings of the 10th International Conference on Learning Representations (ICLR'22), 2022.
  

• Lue Tao, Lei Feng, Jinfeng Yi, Songcan Chen.
  With False Friends Like These, Who Can Notice Mistakes?
  Proceedings of the 36th AAAI Conference on Artificial Intelligence (AAAI'22), pp. 8458-8466, 2022.
  (Acceptance Rate: ~15%)

• Lei Feng, Jun Huang, Senlin Shu, Bo An.
  Regularized Matrix Factorization for Multi-Label Learning with Missing Labels.
  IEEE Transactions on Cybernetics (TCYB), vol. 52, no. 5, pp. 3710-3721, 2022.
  

• Lue Tao, Lei Feng, Jinfeng Yi, Shengjun Huang, Songcan Chen.
  Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training.
  Proceedings of the 35th Conference on Neural Information Processing Systems (NeurIPS'21), pp. 16209-16225, 2021.
  

• Dengbao Wang, Lei Feng, Minling Zhang.
  Rethinking Calibration of Deep Neural Networks: Do Not Be Afraid of Overconfidence.
  Proceedings of the 35th Conference on Neural Information Processing Systems (NeurIPS'21), pp. 11809-11820, 2021.
  

• Lei Feng, Senlin Shu, Yuzhou Cao, Lue Tao, Hongxin Wei, Tao Xiang, Bo An, Gang Niu.
  Multiple-Instance Learning from Similar and Dissimilar Bags.
  Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD'21), pp. 374-382, 2021.
  (Acceptance Rate: 15.4%)

• Lei Feng, Senlin Shu, Nan Lu, Bo Han, Miao Xu, Gang Niu, Bo An, Masashi Sugiyama.
  Pointwise Binary Classification with Pairwise Confidence Comparisons.
  Proceedings of the 38th International Conference on Machine Learning (ICML'21), pp. 3252-3262, 2021.
  

• Yuzhou Cao, Lei Feng, Yitian Xu, Bo An, Gang Niu, Masashi Sugiyama.
  Learning from Similarity-Confidence Data.
  Proceedings of the 38th International Conference on Machine Learning (ICML'21), pp. 1272-1282, 2021.
  

• Lei Feng, Jiaqi Lv, Bo Han, Miao Xu, Gang Niu, Xin Geng, Bo An, Masashi Sugiyama.
  Provably Consistent Partial-Label Learning.
  Proceedings of the 34th Conference on Neural Information Processing Systems (NeurIPS'20), pp. 10948-10960, 2020.
  

• Lei Feng*†, Takuo Kaneko†, Bo Han, Gang Niu, Bo An, Masashi Sugiyama.
  Learning with Multiple Complementary Labels.
  Proceedings of the 37th International Conference on Machine Learning (ICML'20), pp. 3072-3081, 2020.
  

• Jiaqi Lv, Miao Xu, Lei Feng, Gang Niu, Xin Geng, Masashi Sugiyama.
  Progressive Identification of True Labels for Partial-Label Learning.
  Proceedings of the 37th International Conference on Machine Learning (ICML'20), pp. 6500-6510, 2020.
  

• Lei Feng, Senlin Shu, Zhuoyi Lin, Fengmao Lv, Li Li, Bo An.
  Can Cross Entropy Loss Be Robust to Label Noise?
  Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI'20), pp. 2206-2212, 2020.
  (Acceptance Rate: 12.6%)

• Jun Huang*, Linchuan Xu, Jing Wang, Lei Feng*, Kenji Yamanishi.
  Discovering Latent Class Labels for Multi-Label Learning.
  Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI'20), pp. 3058-3064, 2020.
  (Acceptance Rate: 12.6%)

• Hongxin Wei, Lei Feng*, Xiangyu Chen, Bo An.
  Combating Noisy Labels by Agreement: A Joint Training Method with Co-Regularization.
  Proceedings of the 38th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR'20), pp. 13726-13735, 2020.
  

• Lei Feng, Bo An.
  Partial Label Learning by Semantic Difference Maximization.
  Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI'19), pp. 2294-2300, 2019.
  (Acceptance Rate: 17.9%)

• Lei Feng, Bo An, Shuo He.
  Collaboration based Multi-Label Learning.
  Proceedings of the 33rd AAAI Conference on Artificial Intelligence (AAAI'19), pp. 3550-3557, 2019.
  (Acceptance Rate: 16.2%)

• Lei Feng, Bo An.
  Partial Label Learning with Self-Guided Retraining.
  Proceedings of the 33rd AAAI Conference on Artificial Intelligence (AAAI'19), pp. 3542-3549, 2019.
  (Acceptance Rate: 16.2%)

• Lei Feng, Bo An.
  Leveraging Latent Label Distributions for Partial Label Learning.
  Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI'18), pp. 2107-2113, 2018.
  (Acceptance Rate: 20.5%)