Research map / Label-efficient and robust learning / Domain adaptation and generalization
Evolving Domain Generalization via Latent Structure-Aware Sequential Autoencoder
Tiexin Qin, Shiqi Wang, Haoliang Li
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 categorical prior instead of Gaussian prior for category module (ablation) | LSSAE · PowerSupply | accuracy: 70.1 -> 71.1 | Improved |
| use learnable categorical prior instead of fixed Uniform prior (ablation) | LSSAE · PowerSupply | accuracy: 71.0 -> 71.1 | Improved |
| add category-related module (z^v) (ablation) | LSSAE · PowerSupply | accuracy: 70.7 -> 71.1 | Improved |
| add temporal smooth constraint (ablation) | LSSAE · RMNIST | accuracy: 45.5 -> 46.4 | Improved |
| add temporal smooth constraint (ablation) | LSSAE · CalTran | accuracy: 69.9 -> 70.6 | Improved |
| add temporal smooth constraint (ablation) | LSSAE · RMNIST | variance of test accuracy: 3.6 -> 0.4 | Improved |
| add temporal smooth constraint (ablation) | LSSAE · CalTran | variance of test accuracy: 10.0 -> 4.7 | Improved |
| use LSSAE instead of best baseline DIVA | LSSAE · Covariate-shift datasets (avg) | accuracy: 64.2 -> 70.3 | Improved |
| use LSSAE instead of IRM | LSSAE · Concept-shift datasets (avg) | accuracy: 56.8 -> 58.9 | Improved |
| use LSSAE instead of ERM under abrupt label reversal | LSSAE · Sine-C | accuracy: 62.7 -> 60.8 | Hurt |
| use LSSAE instead of DIVA | LSSAE · Circle | accuracy: 67.9 -> 73.8 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DIVA — DiVA: Diverse Visual Feature Aggregation for Deep Metric Learning (ECCV 2020) · better on 2 of 2 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
- Generalizing to Evolving Domains with Latent Structure-Aware Sequential Autoencoder (ICML 2022 · similarity 0.99)
- Enhancing Evolving Domain Generalization through Dynamic Latent Representations (AAAI 2024 · similarity 0.93)
- Weight Diffusion for Future: Learn to Generalize in Non-Stationary Environments (NeurIPS 2024 · similarity 0.89)
- Continuous Domain Generalization (NeurIPS 2025 · similarity 0.88)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.87)
- Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains (ICML 2025 · 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.
