Research map / Label-efficient and robust learning / Test-time adaptation
The Golden Subspace: Where Efficiency Meets Generalization in Continual Test-Time Adaptation
Guannan Lai, Da-Wei Zhou, Zhenguo Li, Han-Jia Ye
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 golden-subspace projection initialized from W^T W (ablation) | GOLD · ImageNet-C | error rate (%): 62.45 -> 60.24 | Improved |
| Add golden-subspace projection initialized from W^T W (ablation) | GOLD · CIFAR10-C | error rate (%): 16.24 -> 15.11 | Improved |
| Add AGOP online update to W^T W initialization (ablation) | GOLD · CIFAR10-C | error rate (%): 15.11 -> 15.06 | No clear effect |
| Add AGOP online update on top of init plus contrastive loss (ablation) | GOLD · CIFAR10-C | error rate (%): 14.64 -> 14.12 | Improved |
| Add AGOP online update on top of init plus contrastive loss (ablation) | GOLD · ImageNet-C | error rate (%): 60.01 -> 59.32 | Improved |
| Add prototype-based contrastive loss (ablation) | GOLD · ImageNet-C | error rate (%): 59.87 -> 59.32 | Improved |
| Add prototype-based contrastive loss (ablation) | GOLD · CIFAR10-C | error rate (%): 15.06 -> 14.12 | Improved |
| Use GOLD instead of OBAO | CTTA · CIFAR10-C | mean error (%): 14.6 -> 14.1 | Improved |
| Use GOLD instead of CoTTA | CTTA · CarlaTTA highway | mIoU: 33.8 -> 34.5 | Improved |
| Use GOLD instead of Source model | CTTA · CarlaTTA clear2rain | mIoU: 71.8 -> 71.0 | Hurt |
| Use GOLD instead of TENT | CTTA · CarlaTTA dynamic | mIoU: 50.8 -> 48.9 | Hurt |
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.
- TENT — Tent: Fully Test-Time Adaptation by Entropy Minimization (ICLR 2021) · better on 4 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
- Test-time Adaptation for Regression by Subspace Alignment (ICLR 2025 · similarity 0.87)
- Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment (NeurIPS 2025 · similarity 0.87)
- Test-time Adaptation in Non-stationary Environments via Adaptive Representation Alignment (NeurIPS 2024 · similarity 0.86)
- Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts (NeurIPS 2026 · similarity 0.86)
- Backpropagation-free Network for 3D Test-time Adaptation (CVPR 2024 · similarity 0.85)
- PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation (NeurIPS 2025 · similarity 0.85)
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.
