Research map / Label-efficient and robust learning / Domain adaptation and generalization
Evolving Standardization for Continual Domain Generalization over Temporal Drift
Mixue Xie, Shuang Li, Longhui Yuan, Chi Harold Liu, Zehui Dai
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 |
|---|---|---|---|
| add stop-gradient sg() in L_stan (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 55.37 -> 93.97 | Improved |
| add stop-gradient in L_ce and L_stan (vs only L_stan) (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 93.97 -> 95.53 | Improved |
| add truncation when sampling features (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 94.46 -> 95.53 | Improved |
| add adversarial and consistency losses (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 93.25 -> 95.53 | Improved |
| add consistency loss L_con (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 94.75 -> 95.53 | Improved |
| add adversarial loss L_adv (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 94.27 -> 95.53 | Improved |
| multi-scale attention instead of single-scale (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 94.09 -> 95.53 | Improved |
| use all historical domain distributions instead of one random (ablation) | EvoS · Yearbook | OOD avg accuracy (%): 95.40 -> 95.53 | Improved |
| use all historical distributions instead of one random (ablation) | EvoS · RMNIST | OOD worst accuracy (%): 85.46 -> 87.04 | Improved |
| use all historical distributions instead of one random (ablation) | EvoS · fMoW | OOD worst accuracy (%): 44.76 -> 45.86 | Improved |
| use all historical distributions instead of one random (ablation) | EvoS · Huffpost | OOD avg accuracy (%): 71.57 -> 72.36 | Improved |
| use all historical distributions instead of one random (ablation) | EvoS · Arxiv | OOD worst accuracy (%): 49.52 -> 50.19 | Improved |
Closest papers by meaning
- Continuous Temporal Domain Generalization (NeurIPS 2024 · similarity 0.90)
- Enhancing Evolving Domain Generalization through Dynamic Latent Representations (AAAI 2024 · similarity 0.90)
- Weight Diffusion for Future: Learn to Generalize in Non-Stationary Environments (NeurIPS 2024 · similarity 0.89)
- Continuous Domain Generalization (NeurIPS 2025 · similarity 0.89)
- Generalizing across Temporal Domains with Koopman Operators (AAAI 2024 · similarity 0.88)
- Improving Domain Generalization with Domain Relations (ICLR 2024 · similarity 0.87)
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.
