real-world datasets
Data collected from real-world scenarios that reflect actual occurrences and behaviors, often used to train and evaluate AI models. They are crucial for ensuring the robustness and generalizability of AI solutions.
- 3D Interaction Geometric Pre-training for Molecular Relational Learning
- A Geometry-Aware Metric for Mode Collapse in Time Series Generative Models
- A Physics-preserved Transfer Learning Method for Differential Equations
- A Unified Framework for Fair Graph Generation: Theoretical Guarantees and Empirical Advances
- Addressing Mark Imbalance in Integration-free Marked Temporal Point Processes
- Attack by Yourself: Effective and Unnoticeable Multi-Category Graph Backdoor Attacks with Subgraph Triggers Pool
- Attention with Trained Embeddings Provably Selects Important Tokens
- AutoDiscovery: Open-ended Scientific Discovery via Bayesian Surprise
- AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning
- CoC-VLA: Delving into Adversarial Domain Transfer for Explainable Autonomous Driving via Chain-of-Causality Visual-Language-Action Model
- Complete Structure Guided Point Cloud Completion via Cluster- and Instance-Level Contrastive Learning
- Conformal Prediction for Causal Effects of Continuous Treatments
- Conformal Prediction for Time-series Forecasting with Change Points
- Constrained Posterior Sampling: Time Series Generation with Hard Constraints
- Coresets for Clustering Under Stochastic Noise
- DBLoss: Decomposition-based Loss Function for Time Series Forecasting
- Deep Continuous-Time State-Space Models for Marked Event Sequences
- Evaluating LLM-contaminated Crowdsourcing Data Without Ground Truth
- Extracting task-relevant preserved dynamics from contrastive aligned neural recordings
- Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference
- Fully Dynamic Algorithms for Chamfer Distance
- HPSERec: A Hierarchical Partitioning and Stepwise Enhancement Framework for Long-tailed Sequential Recommendation
- Hierarchical Demonstration Order Optimization for Many-shot In-Context Learning
- Incentivizing Time-Aware Fairness in Data Sharing
- Individual Fairness In Strategic Classification
- Individually Fair Diversity Maximization
- Influence Functions for Edge Edits in Non-Convex Graph Neural Networks
- Inverse Methods for Missing Data Imputation
- Inverse Optimization Latent Variable Models for Learning Costs Applied to Route Problems
- Is Noise Conditioning Necessary? A Unified Theory of Unconditional Graph Diffusion Models
- Joint‑Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self‑Supervised Learning
- LeapFactual: Reliable Visual Counterfactual Explanation Using Conditional Flow Matching
- Learned Prefix Caching for Efficient LLM Inference
- Learning Across the Gap: Hybrid Multi-armed Bandits with Heterogeneous Offline and Online Data
- Learning-Augmented Streaming Algorithms for Correlation Clustering
- LoSplit: Loss-Guided Dynamic Split for Training-Time Defense Against Graph Backdoor Attacks
- MARS-VFL: A Unified Benchmark for Vertical Federated Learning with Realistic Evaluation
- Magical: Medical Lay Language Generation via Semantic Invariance and Layperson-tailored Adaptation
- Missing Data Imputation by Reducing Mutual Information with Rectified Flows
- Model Editing for Vision Transformers
- Multi-modal contrastive learning adapts to intrinsic dimensions of shared latent variables
- Multivariate Time Series Anomaly Detection with Idempotent Reconstruction
- Noisy Multi-Label Learning through Co-Occurrence-Aware Diffusion
- Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecasting
- On the Value of Cross-Modal Misalignment in Multimodal Representation Learning
- OnlineSplatter: Pose-Free Online 3D Reconstruction for Free-Moving Objects
- Opinion Maximization in Social Networks by Modifying Internal Opinions
- Out-of-Distribution Generalized Graph Anomaly Detection with Homophily-aware Environment Mixup
- Over-squashing in Spatiotemporal Graph Neural Networks
- Path-Enhanced Contrastive Learning for Recommendation
- PoGDiff: Product-of-Gaussians Diffusion Models for Imbalanced Text-to-Image Generation
- PointMAC: Meta-Learned Adaptation for Robust Test-Time Point Cloud Completion
- Practical do-Shapley Explanations with Estimand-Agnostic Causal Inference
- RGB-Only Supervised Camera Parameter Optimization in Dynamic Scenes
- RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
- Redundancy-Aware Test-Time Graph Out-of-Distribution Detection
- Rescaled Influence Functions: Accurate Data Attribution in High Dimension
- Revolutionizing Graph Aggregation: From Suppression to Amplification via BoostGCN
- Robust Neural Rendering in the Wild with Asymmetric Dual 3D Gaussian Splatting
- Robustifying Learning-Augmented Caching Efficiently without Compromising 1-Consistency
- Scalable Cross-View Sample Alignment for Multi-View Clustering with View Structure Similarity
- Scaling Epidemic Inference on Contact Networks: Theory and Algorithms
- Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals
- Semi-supervised Graph Anomaly Detection via Robust Homophily Learning
- ShapeX: Shapelet-Driven Post Hoc Explanations for Time Series Classification Models
- Struct-Bench: A Benchmark for Differentially Private Structured Text Generation
- Timely Clinical Diagnosis through Active Test Selection
- Topology-Aware Conformal Prediction for Stream Networks
- Towards Accurate Time Series Forecasting via Implicit Decoding
- TrajAgent: An LLM-Agent Framework for Trajectory Modeling via Large-and-Small Model Collaboration
- Unlocker: Disentangle the Deadlock of Learning between Label-noisy and Long-tailed Data