domain shift
Domain shift occurs when the statistical properties of training and evaluation datasets differ. It presents challenges for AI models, often resulting in reduced performance when the model is applied to new, unseen data distributions.
- A Generalized Label Shift Perspective for Cross-Domain Gaze Estimation
- A Physics-preserved Transfer Learning Method for Differential Equations
- Buffer layers for Test-Time Adaptation
- Flick: Empowering Federated Learning with Commonsense Knowledge
- OWMM-Agent: Open World Mobile Manipulation With Multi-modal Agentic Data Synthesis
- SPFL: Sequential updates with Parallel aggregation for Enhanced Federated Learning under Category and Domain Shifts
- Towards Single-Source Domain Generalized Object Detection via Causal Visual Prompts