distribution shifts
A phenomenon where the statistical properties of training data differ significantly from those of test data, necessitating adjustments in model training or evaluation to maintain predictive accuracy across varied conditions.
- Agents Robust to Distribution Shifts Learn Causal World Models Even Under Mediation
- Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment
- Beyond Greedy Exits: Improved Early Exit Decisions for Risk Control and Reliability
- Breakthrough Sensor-Limited Single View: Towards Implicit Temporal Dynamics for Time Series Domain Adaptation
- Conformal Prediction under Lévy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations
- D2SA: Dual-Stage Distribution and Slice Adaptation for Efficient Test-Time Adaptation in MRI Reconstruction
- E-BATS: Efficient Backpropagation-Free Test-Time Adaptation for Speech Foundation Models
- Enhancing Deep Batch Active Learning for Regression with Imperfect Data Guided Selection
- Enhancing Visual Prompting through Expanded Transformation Space and Overfitting Mitigation
- Exploring and Leveraging Class Vectors for Classifier Editing
- Exploring the Noise Robustness of Online Conformal Prediction
- FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts
- Failure Prediction at Runtime for Generative Robot Policies
- Feature-Based Instance Neighbor Discovery: Advanced Stable Test-Time Adaptation in Dynamic World
- Geometric Logit Decoupling for Energy-Based Graph Out-of-distribution Detection
- IDOL: Meeting Diverse Distribution Shifts with Prior Physics for Tropical Cyclone Multi-Task Estimation
- Improving Time Series Forecasting via Instance-aware Post-hoc Revision
- Is Limited Participant Diversity Impeding EEG-based Machine Learning?
- Latent Space Factorization in LoRA
- Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift
- MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
- Meta-D2AG: Causal Graph Learning with Interventional Dynamic Data
- Mint: A Simple Test-Time Adaptation of Vision-Language Models against Common Corruptions
- Multi-Expert Distributionally Robust Optimization for Out-of-Distribution Generalization
- Non-exchangeable Conformal Prediction with Optimal Transport: Tackling Distribution Shift with Unlabeled Data
- Online Time Series Forecasting with Theoretical Guarantees
- Precise Diffusion Inversion: Towards Novel Samples and Few-Step Models
- Pruning Spurious Subgraphs for Graph Out-of-Distribution Generalization
- Rethinking Entropy in Test-Time Adaptation: The Missing Piece from Energy Duality
- Sample-Efficient Tabular Self-Play for Offline Robust Reinforcement Learning
- TRUST: Test-Time Refinement using Uncertainty-Guided SSM Traverses
- TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster
- Test-Time Adaptation by Causal Trimming
- The Boundaries of Fair AI in Medical Image Prognosis: A Causal Perspective
- The Rich and the Simple: On the Implicit Bias of Adam and SGD
- Tracing the Roots: Leveraging Temporal Dynamics in Diffusion Trajectories for Origin Attribution
- Uncertainty-Informed Meta Pseudo Labeling for Surrogate Modeling with Limited Labeled Data
- VESSA: Video-based objEct-centric Self-Supervised Adaptation for Visual Foundation Models
- Visual Instruction Bottleneck Tuning
- Weak-to-Strong Generalization under Distribution Shifts
- What Can RL Bring to VLA Generalization? An Empirical Study