Research map / Time series / Causal and generative time series
Diversify: A General Framework for Time Series Out-of-Distribution Detection and Generalization
Lu Wang, Jindong Wang, Xinwei Sun, Yiqiang Chen, Xiangyang Ji, Qiang Yang, Xing Xie
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| use Diversify instead of the second-best baseline (CORAL) | CORAL · EMG OOD detection | ID accuracy (%): 76.50 -> 86.32 | Improved |
| add Diversify with MCP score over the second-best | GroupDRO-MCP · EMG OOD detection | AUROC / AUPR: +4.67% AUROC, +2.33% AUPR | Improved |
| add Diversify with Mahalanobis score over the second-best | CORAL-MAH · EMG OOD detection | AUROC / AUPR: +14.06% AUROC, +3.94% AUPR | Improved |
| replace MCP with Mahalanobis distance as detection score | Diversify · USC-HAD X-Person detection | AUROC: 37.50 -> 99.56 | Improved |
| use Diversify instead of GroupDRO | GroupDRO · WESAD OOD detection | ID accuracy (%): 59.18 -> 64.37 | Improved |
| use Diversify for cross-person generalization on EMG | second-best method · EMG generalization | accuracy (%): +4.3% | Improved |
| use Diversify instead of GroupDRO with a basic CNN | GroupDRO · Speech Commands | accuracy (%): over 1% improvement | Improved |
| use Diversify instead of GroupDRO with MatchboxNet3-1-64 | GroupDRO · Speech Commands | accuracy (%): over 0.6% improvement | Improved |
| use Diversify for generalization on WESAD | other state-of-the-art methods · WESAD generalization | accuracy (%): over 8% | Improved |
| use Diversify for cross-dataset generalization | DANN · HAR cross-dataset generalization | accuracy (%): 41.2 -> 51.1 | Improved |
| use Diversify for one-person-to-another generalization | RSC · HAR one-person-to-another | accuracy (%): 55.9 -> 61.7 | Improved |
| add pseudo domain labels from class-invariant features (step 3) (ablation) | Diversify · EMG | accuracy / AUROC | Improved |
Later papers that compared with this method
1 later accepted paper reports DIVERSIFY as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- CURE: Context-driven Diffusion with Progressive Expansion for Single Domain Generalization in Time Series Classification (ICML 2026 · better on 5 of 5 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Unraveling the Key Components of OOD Generalization via Diversification (ICLR 2024 · similarity 0.86)
- Out-Of-Distribution Detection with Diversification (Provably) (NeurIPS 2024 · similarity 0.85)
- Tri-Level Navigator: LLM-Empowered Tri-Level Learning for Time Series OOD Generalization (NeurIPS 2024 · similarity 0.85)
- Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection (NeurIPS 2026 · similarity 0.84)
- Time Series Generation Under Data Scarcity: A Unified Generative Modeling Approach (NeurIPS 2025 · similarity 0.84)
- Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection (NeurIPS 2024 · 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.
