distributional shifts
Distributional shifts refer to changes in the statistical properties of data encountered during model deployment, crucial for AI systems as they can degrade model performance if not properly addressed.
- $\texttt{AVROBUSTBENCH}$: Benchmarking the Robustness of Audio-Visual Recognition Models at Test-Time
- A Unified Framework for the Transportability of Population-Level Causal Measures
- Active Test-time Vision-Language Navigation
- Aligning Evaluation with Clinical Priorities: Calibration, Label Shift, and Error Costs
- Conformal Prediction Beyond the Horizon: Distribution-Free Inference for Policy Evaluation
- Distributional Adversarial Attacks and Training in Deep Hedging
- Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
- Generalized Category Discovery under Domain Shift: A Frequency Domain Perspective
- HiMoLE: Towards OOD-Robust LoRA via Hierarchical Mixture of Experts
- Imagine Beyond ! Distributionally Robust Autoencoding for State Space Coverage in Online Reinforcement Learning
- Learning Across the Gap: Hybrid Multi-armed Bandits with Heterogeneous Offline and Online Data
- Learning to Specialize: Joint Gating-Expert Training for Adaptive MoEs in Decentralized Settings
- Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecasting
- Leveraging robust optimization for llm alignment under distribution shifts
- LinEAS: End-to-end Learning of Activation Steering with a Distributional Loss
- MetaKoopman: Bayesian Meta-Learning of Koopman Operators for Modeling Structured Dynamics under Distribution Shifts
- Mysteries of the Deep: Role of Intermediate Representations in Out of Distribution Detection
- Omni-Mol: Multitask Molecular Model for Any-to-any Modalities
- Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection
- Quantifying Distributional Invariance in Causal Subgraph for IRM-Free Graph Generalization
- Structural Information-based Hierarchical Diffusion for Offline Reinforcement Learning
- Uncertainty Estimation on Graphs with Structure Informed Stochastic Partial Differential Equations