Research map / Label-efficient and robust learning / Test-time adaptation
Wild-Time: A Benchmark of in-the-Wild Distribution Shift over Time
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Evaluate models on future timestamps rather than held-out training timestamps | ERM across Wild-Time · All six tasks | performance | Hurt |
| Use continual-learning training instead of ERM | arXiv and MIMIC-Readmission · arXiv and MIMIC-Readmission | OOD performance | Helped |
| Use continual-learning training instead of ERM | HuffPost · HuffPost | worst OOD performance | Helped |
| Apply invariant-learning methods instead of ERM | Wild-Time invariant methods · Most Wild-Time tasks | OOD performance | No clear effect |
| Apply invariant-learning methods instead of ERM | Wild-Time invariant methods · Some Wild-Time tasks | OOD performance | Hurt |
| Use self-supervised learning instead of ERM | SimCLR and SwaV · Yearbook and FMoW-Time | OOD performance | No clear effect |
| Use stochastic weight averaging instead of ERM | SWA · Wild-Time tasks | OOD performance | Mixed |
| Use ERM rather than existing robustness approaches | Wild-Time baselines · All six tasks | OOD robustness | No clear effect |
Closest papers by meaning
- "What is Different Between These Datasets?" A Framework for Explaining Data Distribution Shifts (NeurIPS 2026 · similarity 0.86)
- Model Assessment and Selection under Temporal Distribution Shift (ICML 2024 · similarity 0.85)
- Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data (NeurIPS 2024 · similarity 0.85)
- Optimal Aggregation of Prediction Intervals under Unsupervised Domain Shift (NeurIPS 2024 · similarity 0.85)
- Time-Varying Propensity Score to Bridge the Gap between the Past and Present (ICLR 2024 · similarity 0.84)
- Generalization vs Specialization under Concept Shift (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.