Research map / Label-efficient and robust learning / Test-time adaptation
Back to Source: Open-Set Continual Test-Time Adaptation via Domain Compensation
Yingkai Yang, Chaoqi Chen, Hui Huang
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 structural regularizer (R) (ablation) | DOCO w/o S.O · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 64.0 -> 67.6 | Improved |
| add sample splitting and OOD propagation (S, O) (ablation) | DOCO w/o R · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 64.0 -> 68.5 | Improved |
| add structural regularizer on top of S and O (ablation) | DOCO · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 68.5 -> 70.1 | Improved |
| add OOD propagation (O) (ablation) | DOCO w/o O · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 65.9 -> 70.1 | Improved |
| add sample splitting (S) alone (ablation) | prompt-based statistical alignment · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 64.0 -> 65.1 | Improved |
| add S, O and R modules (ablation) | norm-based baseline · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 60.7 -> 68.1 | Improved |
| use prompt learning instead of norm-layer updates (ablation) | DOCO full · ImageNet-C OCTTA (kappa 0.5) | H-score (%): 68.1 -> 70.1 | Improved |
| add structural regularizer in small-batch regime (ablation) | DOCO · ImageNet-C OCTTA (kappa 0.5) | H-score (%) | Improved |
| increase number of prompts L without raising beta (ablation) | DOCO · ImageNet-C OCTTA (kappa 0.5) | H-score (%) | Hurt |
| use DOCO instead of source model in closed-set | Source model · Closed-set CTTA average | ACC (%): 36.9 -> 43.1 | Improved |
| use DOCO instead of best baseline on OCTTA | DPCore · LAION-C OCTTA | H-score (%): 30.3 -> 32.7 | Improved |
| reduce OOD ratio to 0.1 (ablation) | DOCO · ImageNet-C OCTTA | H-score (%): 70.1 -> 69.2 | No clear effect |
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.
- STAMP — STAMP: Outlier-Aware Test-Time Adaptation with Stable Memory Replay (ECCV 2024) · better on 1 of 1 shared result
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
- ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains (NeurIPS 2025 · similarity 0.90)
- DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation (ICML 2025 · similarity 0.89)
- ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation (ICLR 2024 · similarity 0.88)
- Class-aware Domain Knowledge Fusion and Fission for Continual Test-Time Adaptation (NeurIPS 2025 · similarity 0.88)
- A Versatile Framework for Continual Test-Time Domain Adaptation: Balancing Discriminability and Generalizability (CVPR 2024 · similarity 0.88)
- Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts (NeurIPS 2026 · similarity 0.88)
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.
