TPAMI 2025 papers on Label-efficient and robust learning
111 of the 932 papers published in TPAMI 2025 (11.9%) are on Label-efficient and robust learning, compared with 15.6% at TPAMI 2024. They fall into 9 topics.
See TPAMI 2025 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Recognition and long-tail · 43 papers
- Reliable Representation Learning for Incomplete Multi-View Missing Multi-Label Classification (115 citations)
- Partial Multiview Incomplete Multilabel Learning via Uncertainty-Driven Reliable Dynamic Fusion (58 citations)
- Reason and Discovery: A New Paradigm for Open Set Recognition (53 citations)
- OpenGait: A Comprehensive Benchmark Study for Gait Recognition Toward Better Practicality (51 citations)
- Dual-Level Matching With Outlier Filtering for Unsupervised Visible-Infrared Person Re-Identification (42 citations)
- Toward Collaborative Autonomous Driving: Simulation Platform and End-to-End System (39 citations)
- ProtoGCD: Unified and Unbiased Prototype Learning for Generalized Category Discovery (32 citations)
- End-to-End Autonomous Driving Without Costly Modularization and 3D Manual Annotation (30 citations)
- Gait Recognition in the Wild: A Large-Scale Benchmark and NAS-Based Baseline (26 citations)
- BackMix: Regularizing Open Set Recognition by Removing Underlying Fore-Background Priors (21 citations)
- Learning the Optimal Discriminant SVM With Feature Extraction (18 citations)
- Predicting and Enhancing the Fairness of DNNs With the Curvature of Perceptual Manifolds (16 citations)
and 31 more in the interactive map.
Domain adaptation and generalization · 17 papers
- CrossEarth: Geospatial Vision Foundation Model for Domain Generalizable Remote Sensing Semantic Segmentation (47 citations)
- Condition-Invariant Semantic Segmentation (23 citations)
- PhysMLE: Generalizable and Priors-Inclusive Multi-Task Remote Physiological Measurement (23 citations)
- SeCoV2: Semantic Connectivity-Driven Pseudo-Labeling for Robust Cross-Domain Semantic Segmentation (21 citations)
- Unraveling the Mysteries of Label Noise in Source-Free Domain Adaptation: Theory and Practice (18 citations)
- Step-Wise Distribution-Aligned Style Prompt Tuning for Source-Free Cross-Domain Few-Shot Learning (14 citations)
- Hard-Aware Instance Adaptive Self-Training for Unsupervised Cross-Domain Semantic Segmentation (12 citations)
- GLC++: Source-Free Universal Domain Adaptation Through Global-Local Clustering and Contrastive Affinity Learning (11 citations)
- Unified Modality Separation: A Vision-Language Framework for Unsupervised Domain Adaptation (11 citations)
- Beyond Batch Learning: Global Awareness Enhanced Domain Adaptation (7 citations)
- Incorporating Pre-Training Data Matters in Unsupervised Domain Adaptation (4 citations)
- Learning Generalized Medical Image Representation by Decoupled Feature Queries (4 citations)
and 5 more in the interactive map.
Semi-supervised learning · 12 papers
- UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation (142 citations)
- Robust Semi-Supervised Feature Selection With Multi-Granularity Zentropy Modeling (24 citations)
- Semantic-Aware Pseudo-Labeling for Unsupervised Meta-Learning (19 citations)
- Semi-Supervised Counting via Pixel-by-Pixel Density Distribution Modeling (16 citations)
- Semi-Supervised Risk Control via Prediction-Powered Inference (14 citations)
- Towards a Theoretical Understanding of Semi-Supervised Learning Under Class Distribution Mismatch (5 citations)
- An Enhanced Adaptive Confidence Margin for Semi-Supervised Facial Expression Recognition (4 citations)
- ClusMatch: Improving Deep Clustering by Unified Positive and Negative Pseudo-Label Learning (3 citations)
- Contrastive Learning for Semi-Supervised Deep Regression With Generalized Ordinal Rankings From Spectral Seriation (3 citations)
- SEMI-CAVA: A Causal Variational Approach to Semi-Supervised Learning (3 citations)
- Variational Bayesian Semi-Supervised Keyword Extraction (1 citation)
- Generalized Conditional Similarity Learning via Semantic Matching
Self-supervised and contrastive · 12 papers
- Large-Scale 3D Medical Image Pre-Training With Geometric Context Priors (38 citations)
- Bootstrap Masked Visual Modeling via Hard Patch Mining (10 citations)
- CCDPlus: Towards Accurate Character to Character Distillation for Text Recognition (9 citations)
- Neural Eigenfunctions are Structured Representation Learners (9 citations)
- Contrastive Learning via Variational Information Bottleneck (8 citations)
- RF-URL 2.0: A General Unsupervised Representation Learning Method for RF Sensing (8 citations)
- Enhancing Representations Through Heterogeneous Self-Supervised Learning (5 citations)
- Semantic Concentration for Self-Supervised Dense Representations Learning (3 citations)
- Weakly and Self-Supervised Class-Agnostic Motion Prediction for Autonomous Driving (3 citations)
- Neural Prediction Errors as a Unified Cue for Abstract Visual Reasoning
- NVS-SQA: Exploring Self-Supervised Quality Representation Learning for Neurally Synthesized Scenes Without References
- Self-Supervised Skeleton Representation Learning Via Actionlet Contrast and Reconstruct
OOD and anomaly detection · 11 papers
- M3DM-NR: RGB-3D Noisy-Resistant Industrial Anomaly Detection via Multimodal Denoising (37 citations)
- Identifying Semantic Component for Robust Molecular Property Prediction (19 citations)
- Mixture of Gaussian-Distributed Prototypes With Generative Modelling for Interpretable and Trustworthy Image Recognition (16 citations)
- Model-Free Test Time Adaptation for Out-of-Distribution Detection (10 citations)
- W-DOE: Wasserstein Distribution-Agnostic Outlier Exposure (9 citations)
- InfoBound: A Provable Information-Bounds Inspired Framework for Both OoD Generalization and OoD Detection (5 citations)
- Enhanced Dual-Pattern Matching With Vision-Language Representation for Out-of-Distribution Detection (4 citations)
- A Unified Decision Rule for Generalized Out-of-Distribution Detection (3 citations)
- LMP-GAN: Out-of-Distribution Detection for Non-Control Data Malware Attacks (3 citations)
- Unknown-Aware Bilateral Dependency Optimization for Defending Against Model Inversion Attacks (3 citations)
- Distilling the Unknown to Unveil Certainty (1 citation)
Knowledge distillation · 8 papers
- CMKD: CNN/Transformer-Based Cross-Model Knowledge Distillation for Audio Classification (41 citations)
- DIST+: Knowledge Distillation From a Stronger Adaptive Teacher (30 citations)
- CLEAN: Category Knowledge-Driven Compression Framework for Efficient 3D Object Detection (6 citations)
- Accelerated Self-Supervised Multi-Illumination Color Constancy With Hybrid Knowledge Distillation (2 citations)
- Toward Effective Knowledge Distillation: Navigating Beyond Small-Data Pitfall (2 citations)
- Continuous Review and Timely Correction: Enhancing the Resistance to Noisy Labels via Self-Not-True and Class-Wise Distillation (1 citation)
- Forget Me Not: Fighting Local Overfitting With Knowledge Fusion and Distillation (1 citation)
- Aligning Logits Generatively for Principled Black-Box Knowledge Distillation in the Wild
Noisy labels · 5 papers
- Reliable Few-Shot Learning Under Dual Noises (42 citations)
- IRNet: Iterative Refinement Network for Noisy Partial Label Learning (12 citations)
- Jo-SNC: Combating Noisy Labels Through Fostering Self- and Neighbor-Consistency (7 citations)
- A Theoretical Perspective on Streaming Noisy Data With Distribution Shift (1 citation)
- Improving the Instance-Dependent Transition Matrix Estimation by Exploiting Self-Supervised Learning (1 citation)
