AAAI 2024 accepted papers on Label-efficient and robust learning

Recognition and long-tail · 206 papers

• Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Domain Learning (193 citations)
• SkeletonGait: Gait Recognition Using Skeleton Maps (80 citations)
• High-Order Structure Based Middle-Feature Learning for Visible-Infrared Person Re-identification (58 citations)
• Feature Fusion from Head to Tail for Long-Tailed Visual Recognition (52 citations)
• SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentation (51 citations)
• DMMR: Cross-Subject Domain Generalization for EEG-Based Emotion Recognition via Denoising Mixed Mutual Reconstruction (42 citations)
• Distribution Matching for Multi-Task Learning of Classification Tasks: A Large-Scale Study on Faces & Beyond (39 citations)
• Attention-Induced Embedding Imputation for Incomplete Multi-View Partial Multi-Label Classification (38 citations)
• Deep Unfolded Network with Intrinsic Supervision for Pan-Sharpening (37 citations)
• Multi-Label Supervised Contrastive Learning (35 citations)
• Learning Accurate and Bidirectional Transformation via Dynamic Embedding Transportation for Cross-Domain Recommendation (34 citations)
• Occluded Person Re-identification via Saliency-Guided Patch Transfer (34 citations)

and 194 more in the interactive map.

Domain adaptation and generalization · 57 papers

• Prompt-Based Distribution Alignment for Unsupervised Domain Adaptation (76 citations)
• Graph Disentangled Contrastive Learning with Personalized Transfer for Cross-Domain Recommendation (32 citations)
• Learning Generalized Medical Image Segmentation from Decoupled Feature Queries (31 citations)
• Learning Generalized Segmentation for Foggy-Scenes by Bi-directional Wavelet Guidance (31 citations)
• Effective Comparative Prototype Hashing for Unsupervised Domain Adaptation (27 citations)
• Model Reprogramming: Resource-Efficient Cross-Domain Machine Learning (27 citations)
• AdvST: Revisiting Data Augmentations for Single Domain Generalization (25 citations)
• Mean Teacher DETR with Masked Feature Alignment: A Robust Domain Adaptive Detection Transformer Framework (21 citations)
• Agile Multi-Source-Free Domain Adaptation (18 citations)
• Cycle Self-Refinement for Multi-Source Domain Adaptation (18 citations)
• MLNet: Mutual Learning Network with Neighborhood Invariance for Universal Domain Adaptation (18 citations)
• Generalizable Sleep Staging via Multi-Level Domain Alignment (17 citations)

and 45 more in the interactive map.

Zero-/few-shot and meta-learning · 29 papers

• Cross-Layer and Cross-Sample Feature Optimization Network for Few-Shot Fine-Grained Image Classification (55 citations)
• Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition (26 citations)
• Strong Baselines for Parameter-Efficient Few-Shot Fine-Tuning (26 citations)
• Zero-1-to-3: Domain-Level Zero-Shot Cognitive Diagnosis via One Batch of Early-Bird Students towards Three Diagnostic Objectives (24 citations)
• Data-Free Generalized Zero-Shot Learning (18 citations)
• StyleSinger: Style Transfer for Out-of-Domain Singing Voice Synthesis (18 citations)
• Boosting Few-Shot Learning via Attentive Feature Regularization (17 citations)
• Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning (17 citations)
• Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (16 citations)
• Task-Adaptive Prompted Transformer for Cross-Domain Few-Shot Learning (16 citations)
• Collaborative Consortium of Foundation Models for Open-World Few-Shot Learning (15 citations)
• Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (15 citations)

and 17 more in the interactive map.

Semi-supervised learning · 22 papers

• Pseudo-Label Calibration Semi-supervised Multi-Modal Entity Alignment (22 citations)
• RPSC: Robust Pseudo-Labeling for Semantic Clustering (19 citations)
• ANEDL: Adaptive Negative Evidential Deep Learning for Open-Set Semi-supervised Learning (13 citations)
• Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation (12 citations)
• Multilevel Attention Network with Semi-supervised Domain Adaptation for Drug-Target Prediction (12 citations)
• Roll with the Punches: Expansion and Shrinkage of Soft Label Selection for Semi-supervised Fine-Grained Learning (12 citations)
• Data Augmented Graph Neural Networks for Personality Detection (10 citations)
• Semi-supervised Class-Agnostic Motion Prediction with Pseudo Label Regeneration and BEVMix (10 citations)
• Three Heads Are Better than One: Complementary Experts for Long-Tailed Semi-supervised Learning (10 citations)
• BaCon: Boosting Imbalanced Semi-supervised Learning via Balanced Feature-Level Contrastive Learning (9 citations)
• MCSSME: Multi-Task Contrastive Learning for Semi-supervised Singing Melody Extraction from Polyphonic Music (9 citations)
• Unknown-Aware Graph Regularization for Robust Semi-supervised Learning from Uncurated Data (9 citations)

and 10 more in the interactive map.

OOD and anomaly detection · 22 papers

• LogFormer: A Pre-train and Tuning Pipeline for Log Anomaly Detection (70 citations)
• Debiasing Multimodal Sarcasm Detection with Contrastive Learning (35 citations)
• Generating and Reweighting Dense Contrastive Patterns for Unsupervised Anomaly Detection (24 citations)
• Out-of-Distribution Detection in Long-Tailed Recognition with Calibrated Outlier Class Learning (18 citations)
• Toward More Generalized Malicious URL Detection Models (17 citations)
• How to Overcome Curse-of-Dimensionality for Out-of-Distribution Detection? (12 citations)
• Unsupervised Layer-Wise Score Aggregation for Textual OOD Detection (12 citations)
• Dense Projection for Anomaly Detection (11 citations)
• EAT: Towards Long-Tailed Out-of-Distribution Detection (9 citations)
• Learning by Erasing: Conditional Entropy Based Transferable Out-of-Distribution Detection (9 citations)
• TagFog: Textual Anchor Guidance and Fake Outlier Generation for Visual Out-of-Distribution Detection (7 citations)
• Exploring Channel-Aware Typical Features for Out-of-Distribution Detection (6 citations)

and 10 more in the interactive map.

Noisy labels · 20 papers

• Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (32 citations)
• Dual Self-Paced Cross-Modal Hashing (28 citations)
• Toward Robustness in Multi-Label Classification: A Data Augmentation Strategy against Imbalance and Noise (26 citations)
• Negative Pre-aware for Noisy Cross-Modal Matching (18 citations)
• Dirichlet-Based Prediction Calibration for Learning with Noisy Labels (17 citations)
• Unlocking the Power of Open Set: A New Perspective for Open-Set Noisy Label Learning (14 citations)
• Which Is More Effective in Label Noise Cleaning, Correction or Filtering? (14 citations)
• Learning with Noisy Labels Using Hyperspherical Margin Weighting (12 citations)
• Regroup Median Loss for Combating Label Noise (12 citations)
• Coupled Confusion Correction: Learning from Crowds with Sparse Annotations (11 citations)
• Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation (11 citations)
• Dual-Level Curriculum Meta-Learning for Noisy Few-Shot Learning Tasks (8 citations)

and 8 more in the interactive map.

Knowledge distillation · 19 papers

• Understanding the Role of the Projector in Knowledge Distillation (37 citations)
• Boosting Neural Cognitive Diagnosis with Student’s Affective State Modeling (26 citations)
• Layer Attack Unlearning: Fast and Accurate Machine Unlearning via Layer Level Attack and Knowledge Distillation (19 citations)
• Fine-Grained Distillation for Long Document Retrieval (12 citations)
• Self-Training Based Few-Shot Node Classification by Knowledge Distillation (11 citations)
• Discrepancy and Uncertainty Aware Denoising Knowledge Distillation for Zero-Shot Cross-Lingual Named Entity Recognition (8 citations)
• Progressive Distillation Based on Masked Generation Feature Method for Knowledge Graph Completion (8 citations)
• Impartial Adversarial Distillation: Addressing Biased Data-Free Knowledge Distillation via Adaptive Constrained Optimization (7 citations)
• Tail-STEAK: Improve Friend Recommendation for Tail Users via Self-Training Enhanced Knowledge Distillation (7 citations)
• Teacher as a Lenient Expert: Teacher-Agnostic Data-Free Knowledge Distillation (6 citations)
• Boosting Residual Networks with Group Knowledge (5 citations)
• Cooperative Knowledge Distillation: A Learner Agnostic Approach (5 citations)

and 7 more in the interactive map.

Self-supervised and contrastive · 16 papers

• SwitchTab: Switched Autoencoders Are Effective Tabular Learners (30 citations)
• SCD-Net: Spatiotemporal Clues Disentanglement Network for Self-Supervised Skeleton-Based Action Recognition (24 citations)
• Contrastive Tuning: A Little Help to Make Masked Autoencoders Forget (13 citations)
• Partial Multi-View Clustering via Self-Supervised Network (11 citations)
• PPEA-Depth: Progressive Parameter-Efficient Adaptation for Self-Supervised Monocular Depth Estimation (8 citations)
• Measuring Self-Supervised Representation Quality for Downstream Classification Using Discriminative Features (7 citations)
• Quantile-Based Maximum Likelihood Training for Outlier Detection (7 citations)
• DanceMVP: Self-Supervised Learning for Multi-Task Primitive-Based Dance Performance Assessment via Transformer Text Prompting (6 citations)
• DenoSent: A Denoising Objective for Self-Supervised Sentence Representation Learning (6 citations)
• Self-Supervised Representation Learning with Meta Comprehensive Regularization (6 citations)
• Self-Supervised Disentangled Representation Learning for Robust Target Speech Extraction (5 citations)
• DUEL: Duplicate Elimination on Active Memory for Self-Supervised Class-Imbalanced Learning (4 citations)

and 4 more in the interactive map.

Test-time adaptation · 8 papers

• Test-Time Domain Adaptation by Learning Domain-Aware Batch Normalization (24 citations)
• Exploring Sparse Visual Prompt for Domain Adaptive Dense Prediction (23 citations)
• Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization (20 citations)
• Heterogeneous Test-Time Training for Multi-Modal Person Re-identification (17 citations)
• Robust Test-Time Adaptation for Zero-Shot Prompt Tuning (9 citations)
• Test-Time Personalization with Meta Prompt for Gaze Estimation (9 citations)
• Unraveling Batch Normalization for Realistic Test-Time Adaptation (3 citations)
• Test-Time Adaptation via Style and Structure Guidance for Histological Image Registration (2 citations)