representation learning
A set of techniques in machine learning that aims to automatically discover representations of data that make it easier for algorithms to perform tasks, such as classification or regression. In AI, effective representation learning enables models to generalize better and understand the underlying structure of complex data.
- $\textit{HiMaCon:}$ Discovering Hierarchical Manipulation Concepts from Unlabeled Multi-Modal Data
- Act to See, See to Act: Diffusion-Driven Perception-Action Interplay for Adaptive Policies
- Adaptive Gradient Masking for Balancing ID and MLLM-based Representations in Recommendation
- Angular Constraint Embedding via SpherePair Loss for Constrained Clustering
- Automatic Synthetic Data and Fine-grained Adaptive Feature Alignment for Composed Person Retrieval
- Beyond Value Functions: Single-Loop Bilevel Optimization under Flatness Conditions
- BiggerGait: Unlocking Gait Recognition with Layer-wise Representations from Large Vision Models
- Boosting Generative Image Modeling via Joint Image-Feature Synthesis
- CG-SSL: Concept-Guided Self-Supervised Learning
- CORAL: Disentangling Latent Representations in Long-Tailed Diffusion
- Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models
- Connecting Jensen–Shannon and Kullback–Leibler Divergences: A New Bound for Representation Learning
- Contrastive Learning with Data Misalignment: Feature Purity, Training Dynamics and Theoretical Generalization Guarantees
- Contrastive Representations for Temporal Reasoning
- Contrastive Self-Supervised Learning As Neural Manifold Packing
- Copresheaf Topological Neural Networks: A Generalized Deep Learning Framework
- Defining and Discovering Hyper-meta-paths for Heterogeneous Hypergraphs
- Enhancing Text-to-Image Diffusion Transformer via Split-Text Conditioning
- FAME: Adaptive Functional Attention with Expert Routing for Function-on-Function Regression
- Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference
- GreenHyperSpectra: A multi-source hyperspectral dataset for global vegetation trait prediction
- HiPoNet: A Multi-View Simplicial Complex Network for High Dimensional Point-Cloud and Single-Cell data
- Hybrid-Collaborative Augmentation and Contrastive Sample Adaptive-Differential Awareness for Robust Attributed Graph Clustering
- Integrating Drug Substructures and Longitudinal Electronic Health Records for Personalized Drug Recommendation
- Intermediate Domain Alignment and Morphology Analogy for Patent-Product Image Retrieval
- Interpretable and Parameter Efficient Graph Neural Additive Models with Random Fourier Features
- JADE: Joint Alignment and Deep Embedding for Multi-Slice Spatial Transcriptomics
- Latent Zoning Network: A Unified Principle for Generative Modeling, Representation Learning, and Classification
- Learning Diffusion Models with Flexible Representation Guidance
- Learning Without Augmenting: Unsupervised Time Series Representation Learning via Frame Projections
- Long-Tailed Recognition via Information-Preservable Two-Stage Learning
- Looking Beyond the Known: Towards a Data Discovery Guided Open-World Object Detection
- MedicalNarratives: Connecting Medical Vision and Language with Localized Narratives
- Minimal Semantic Sufficiency Meets Unsupervised Domain Generalization
- Mitigating Spurious Features in Contrastive Learning with Spectral Regularization
- NAVIX: Scaling MiniGrid Environments with JAX
- Neural Thermodynamics: Entropic Forces in Deep and Universal Representation Learning
- OLinear: A Linear Model for Time Series Forecasting in Orthogonally Transformed Domain
- OmniSegmentor: A Flexible Multi-Modal Learning Framework for Semantic Segmentation
- On the creation of narrow AI: hierarchy and nonlocality of neural network skills
- Predictive Coding Enhances Meta-RL To Achieve Interpretable Bayes-Optimal Belief Representation Under Partial Observability
- Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning
- Provably Efficient Multi-Task Meta Bandit Learning via Shared Representations
- Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detection
- Rethinking Tokenized Graph Transformers for Node Classification
- Revisiting Orbital Minimization Method for Neural Operator Decomposition
- Revisiting Residual Connections: Orthogonal Updates for Stable and Efficient Deep Networks
- Rotary Masked Autoencoders are Versatile Learners
- Scalable Cross-View Sample Alignment for Multi-View Clustering with View Structure Similarity
- ScatterAD: Temporal-Topological Scattering Mechanism for Time Series Anomaly Detection
- Self-Supervised Learning of Graph Representations for Network Intrusion Detection
- Siegel Neural Networks
- Simple and Efficient Heterogeneous Temporal Graph Neural Network
- Switchable Token-Specific Codebook Quantization For Face Image Compression
- TopER: Topological Embeddings in Graph Representation Learning
- Toward Artificial Palpation: Representation Learning of Touch on Soft Bodies
- Transferring Linear Features Across Language Models With Model Stitching
- UniHG: A Large-scale Universal Heterogeneous Graph Dataset and Benchmark for Representation Learning and Cross-Domain Transferring
- Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains
- VisDiff: SDF-Guided Polygon Generation for Visibility Reconstruction, Characterization and Recognition
- Visual Instruction Bottleneck Tuning