Lei Feng (NTU / RIKEN)


Code and Data


  • Leverage Latent Label Distributions for Partial Label Learning [IJCAI'18]. [LALO]

  • Partial Label Learning with Self-Guided Retraining [AAAI'19]. [SURE]

  • Progressive Identification of True Labels for Partial-Label Learning [ICML'20]. [PRODEN]

  • Combating Noisy Labels by Agreement: A Joint Training Method with Co-Regularization [CVPR'20]. [JoCoR]

  • Embedding-Augmented Generalized Matrix Factorization for Recommendation with Implicit Feedback [IEEE IS'21]. [AGMF]

  • Can Cross Entropy Loss Be Robust to Label Noise [IJCAI'20]. [TCE]

  • Learning with Multiple Complementary Labels [ICML'20]. [LMCL]

  • Provably Consistent Partial-Label Learning [NeurIPS'20]. [RC & CC]

  • Learning from Complementary Labels via Partial-Output Consistency Regularization [IJCAI'21]. [CLL_POCR]

  • Pointwise Binary Classification with Pairwise Confidence Comparisons [ICML'21]. [Pcomp]

  • Learning from Similarity-Confidence Data [ICML'21]. [Sconf]

  • Multiple-Instance Learning from Similar and Dissimilar Bags [KDD'21]. [SDMIL]

  • Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training [NeurIPS'21]. [Delusive-Adversary]

  • With False Friends Like These, Who Can Notice Mistakes [AAAI'22]. [False-Friends]

  • PiCO: Contrastive Label Disambiguation for Partial-Label Learning [ICLR'22]. [PiCO]

  • Exploiting Class Activation Value for Partial-Label Learning [ICLR'22]. [CAVL]

  • Mitigating Neural Network Overconfidence with Logit Normalization [ICML'22]. [LogitNorm]

  • Open-Sampling: Exploring Out-of-Distribution Data for Re-balancing Long-tailed Datasets [ICML'22]. [Open-sampling]

  • SemiNLL: A Framework of Noisy-Label Learning by Semi-Supervised Learning [TMLR'22]. [SemiNLL]

  • SoLar: Sinkhorn Label Refinery for Imbalanced Partial-Label Learning [NeurIPS'22]. [SoLar]

  • Generalizing Consistent Multi-Class Classification with Rejection to be Compatible with Arbitrary Losses [NeurIPS'22]. [CwR]

  • Partial-Label Regression [AAAI'23]. [PLR]

  • A Universal Unbiased Method for Classification from Aggregate Observations [ICML'23]. [UUM]

  • Weakly Supervised Regression with Interval Targets [ICML'23]. [WSRIT]

  • Mitigating Memorization of Noisy Labels by Clipping the Model Prediction [ICML'23]. [LogitClip]

  • Consistent Complementary-Label Learning via Order-Preserving Losses [AISTATS'23]. [OPCLL]

  • Late Stopping: Avoiding Confidently Learning from Mislabeled Examples [ICCV'23]. [LateStopping]

  • Binary Classification with Confidence Difference [NeurIPS'23]. [ConfDiff]

  • Regression with Cost-based Rejection [NeurIPS'23]. [RCR]

  • On the Vulnerability of Adversarially Trained Models Against Two-faced Attacks [ICLR'24]. [TF-Attacks]

  • Consistent Multi-Class Classification from Multiple Unlabeled Datasets [ICLR'24]. [RCM-CCM]

  • Candidate Label Set Pruning: A Data-centric Perspective for Deep Partial-label Learning [ICLR'24]. [CLSP]

  • Early Stopping Against Label Noise Without Validation Data [ICLR'24]. [LabelWave]

  • CroSel: Cross Selection of Confident Pseudo Labels for Partial-Label Learning [CVPR'24]. [CroSel]

  • Exploiting Human-AI Dependency for Learning to Defer [ICML'24]. [DCE]

  • Candidate Pseudolabel Learning: Enhancing Vision-Language Models by Prompt Tuning with Unlabeled Data. [ICML'24]. [CPL]