NeurIPS 2023 accepted papers on Label-efficient and robust learning
342 of the 3,584 papers accepted at NeurIPS 2023 (9.5%) are on Label-efficient and robust learning, compared with 11.9% at NeurIPS 2022. They fall into 9 topics.
See NeurIPS 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Recognition and long-tail · 147 papers
- Recommender Systems with Generative Retrieval (91 citations)
- Realistic Synthetic Financial Transactions for Anti-Money Laundering Models (50 citations)
- An Empirical Investigation of the Role of Pre-training in Lifelong Learning (40 citations)
- rPPG-Toolbox: Deep Remote PPG Toolbox (34 citations)
- IEBins: Iterative Elastic Bins for Monocular Depth Estimation (21 citations)
- Plug-and-Play Stability for Intracortical Brain-Computer Interfaces: A One-Year Demonstration of Seamless Brain-to-Text Communication (19 citations)
- Data Selection for Language Models via Importance Resampling (16 citations)
- Learning to Tokenize for Generative Retrieval (16 citations)
- Online Map Vectorization for Autonomous Driving: A Rasterization Perspective (16 citations)
- Class-Conditional Conformal Prediction with Many Classes (15 citations)
- The Rise of AI Language Pathologists: Exploring Two-level Prompt Learning for Few-shot Weakly-supervised Whole Slide Image Classification (14 citations)
- How Re-sampling Helps for Long-Tail Learning? (13 citations)
and 135 more in the interactive map.
Self-supervised and contrastive · 40 papers
- CROMA: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoders (32 citations)
- Battle of the Backbones: A Large-Scale Comparison of Pretrained Models across Computer Vision Tasks (26 citations)
- Self-Supervised Learning of Representations for Space Generates Multi-Modular Grid Cells (26 citations)
- Cross-Scale MAE: A Tale of Multiscale Exploitation in Remote Sensing (11 citations)
- Reverse Engineering Self-Supervised Learning (11 citations)
- DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning (6 citations)
- Learning Efficient Coding of Natural Images with Maximum Manifold Capacity Representations (5 citations)
- Augmentation-Aware Self-Supervision for Data-Efficient GAN Training (4 citations)
- Greatness in Simplicity: Unified Self-Cycle Consistency for Parser-Free Virtual Try-On (4 citations)
- Self-Weighted Contrastive Learning among Multiple Views for Mitigating Representation Degeneration (4 citations)
- Contextually Affinitive Neighborhood Refinery for Deep Clustering (3 citations)
- Selectivity Drives Productivity: Efficient Dataset Pruning for Enhanced Transfer Learning (3 citations)
and 28 more in the interactive map.
OOD and anomaly detection · 31 papers
- ReContrast: Domain-Specific Anomaly Detection via Contrastive Reconstruction (27 citations)
- Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detection (22 citations)
- Benchmarking Distribution Shift in Tabular Data with TableShift (12 citations)
- LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learning (12 citations)
- RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions (11 citations)
- Learning to Augment Distributions for Out-of-distribution Detection (8 citations)
- Neural Relation Graph: A Unified Framework for Identifying Label Noise and Outlier Data (8 citations)
- GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients (7 citations)
- ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets (7 citations)
- Invariant Learning via Probability of Sufficient and Necessary Causes (6 citations)
- ADGym: Design Choices for Deep Anomaly Detection (5 citations)
- Diversified Outlier Exposure for Out-of-Distribution Detection via Informative Extrapolation (4 citations)
and 19 more in the interactive map.
Semi-supervised learning · 27 papers
- Rethinking Semi-Supervised Medical Image Segmentation: A Variance-Reduction Perspective (62 citations)
- Switching Temporary Teachers for Semi-Supervised Semantic Segmentation (16 citations)
- Towards Generic Semi-Supervised Framework for Volumetric Medical Image Segmentation (16 citations)
- Jaccard Metric Losses: Optimizing the Jaccard Index with Soft Labels (14 citations)
- Harnessing Hard Mixed Samples with Decoupled Regularizer (13 citations)
- FlatMatch: Bridging Labeled Data and Unlabeled Data with Cross-Sharpness for Semi-Supervised Learning (11 citations)
- DAW: Exploring the Better Weighting Function for Semi-supervised Semantic Segmentation (10 citations)
- Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels (9 citations)
- All Points Matter: Entropy-Regularized Distribution Alignment for Weakly-supervised 3D Segmentation (6 citations)
- Dual Mean-Teacher: An Unbiased Semi-Supervised Framework for Audio-Visual Source Localization (6 citations)
- Class-Distribution-Aware Pseudo-Labeling for Semi-Supervised Multi-Label Learning (5 citations)
- InstanT: Semi-supervised Learning with Instance-dependent Thresholds (5 citations)
and 15 more in the interactive map.
Zero-/few-shot and meta-learning · 24 papers
- EHRSHOT: An EHR Benchmark for Few-Shot Evaluation of Foundation Models (21 citations)
- BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecasting (12 citations)
- Zero-Shot Anomaly Detection via Batch Normalization (12 citations)
- DiffKendall: A Novel Approach for Few-Shot Learning with Differentiable Kendall's Rank Correlation (10 citations)
- DrugCLIP: Contrastive Protein-Molecule Representation Learning for Virtual Screening (8 citations)
- CWCL: Cross-Modal Transfer with Continuously Weighted Contrastive Loss (5 citations)
- ProtoDiff: Learning to Learn Prototypical Networks by Task-Guided Diffusion (5 citations)
- Embroid: Unsupervised Prediction Smoothing Can Improve Few-Shot Classification (4 citations)
- Episodic Multi-Task Learning with Heterogeneous Neural Processes (4 citations)
- What a MESS: Multi-Domain Evaluation of Zero-Shot Semantic Segmentation (4 citations)
- Enhancing CLIP with CLIP: Exploring Pseudolabeling for Limited-Label Prompt Tuning (3 citations)
- Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Models (3 citations)
and 12 more in the interactive map.
Domain adaptation and generalization · 24 papers
- MADG: Margin-based Adversarial Learning for Domain Generalization (26 citations)
- Multi-Prompt Alignment for Multi-Source Unsupervised Domain Adaptation (15 citations)
- DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining (14 citations)
- Subspace Identification for Multi-Source Domain Adaptation (12 citations)
- Domain Watermark: Effective and Harmless Dataset Copyright Protection is Closed at Hand (11 citations)
- Make the U in UDA Matter: Invariant Consistency Learning for Unsupervised Domain Adaptation (8 citations)
- Domain Re-Modulation for Few-Shot Generative Domain Adaptation (6 citations)
- Learning Domain-Aware Detection Head with Prompt Tuning (6 citations)
- An Iterative Self-Learning Framework for Medical Domain Generalization (4 citations)
- Uncertainty-Aware Alignment Network for Cross-Domain Video-Text Retrieval (4 citations)
- Diffusion-Based Probabilistic Uncertainty Estimation for Active Domain Adaptation (3 citations)
- Scalarization for Multi-Task and Multi-Domain Learning at Scale (3 citations)
and 12 more in the interactive map.
Noisy labels · 17 papers
- Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels (8 citations)
- AQuA: A Benchmarking Tool for Label Quality Assessment (5 citations)
- Active Negative Loss Functions for Learning with Noisy Labels (3 citations)
- A Path to Simpler Models Starts With Noise (2 citations)
- ALIM: Adjusting Label Importance Mechanism for Noisy Partial Label Learning (1 citation)
- Automated Classification of Model Errors on ImageNet (1 citation)
- CAPro: Webly Supervised Learning with Cross-modality Aligned Prototypes (1 citation)
- [Re] Pure Noise to the Rescue of Insufficient Data
- Calibrating “Cheap Signals” in Peer Review without a Prior
- CS-Isolate: Extracting Hard Confident Examples by Content and Style Isolation
- Label Correction of Crowdsourced Noisy Annotations with an Instance-Dependent Noise Transition Model
- R-divergence for Estimating Model-oriented Distribution Discrepancy
and 5 more in the interactive map.
Knowledge distillation · 17 papers
- One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillation (31 citations)
- KD-Zero: Evolving Knowledge Distiller for Any Teacher-Student Pairs (17 citations)
- XES3G5M: A Knowledge Tracing Benchmark Dataset with Auxiliary Information (10 citations)
- What Knowledge Gets Distilled in Knowledge Distillation? (6 citations)
- Sequential Subset Matching for Dataset Distillation (5 citations)
- Module-wise Adaptive Distillation for Multimodality Foundation Models (4 citations)
- Frequency Domain-Based Dataset Distillation (2 citations)
- Progressive Ensemble Distillation: Building Ensembles for Efficient Inference (2 citations)
- PTADisc: A Cross-Course Dataset Supporting Personalized Learning in Cold-Start Scenarios (2 citations)
- Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering (1 citation)
- L2T-DLN: Learning to Teach with Dynamic Loss Network (1 citation)
- RD-Suite: A Benchmark for Ranking Distillation (1 citation)
and 5 more in the interactive map.
Test-time adaptation · 15 papers
- Align Your Prompts: Test-Time Prompting with Distribution Alignment for Zero-Shot Generalization (14 citations)
- SwapPrompt: Test-Time Prompt Adaptation for Vision-Language Models (14 citations)
- SoTTA: Robust Test-Time Adaptation on Noisy Data Streams (13 citations)
- Learn to Categorize or Categorize to Learn? Self-Coding for Generalized Category Discovery (10 citations)
- MMD Aggregated Two-Sample Test (10 citations)
- Can You Rely on Your Model Evaluation? Improving Model Evaluation with Synthetic Test Data (5 citations)
- Learning to Receive Help: Intervention-Aware Concept Embedding Models (3 citations)
- Neural Priming for Sample-Efficient Adaptation (3 citations)
- Test-Time Amendment with a Coarse Classifier for Fine-Grained Classification (3 citations)
- CODA: Generalizing to Open and Unseen Domains with Compaction and Disambiguation (2 citations)
- Convolution Monge Mapping Normalization for learning on sleep data (2 citations)
- Convolutional Visual Prompt for Robust Visual Perception (2 citations)
and 3 more in the interactive map.