Research map / Continual learning / Domain and prompt-based
Continual Test-Time Domain Adaptation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 CoTTA instead of TENT-online | WideResNet-28 · CIFAR10-to-CIFAR10C, standard | mean error rate (%): 18.6 -> 16.2 | Helped |
| Use CoTTA instead of BN Adapt | WideResNet-28 · CIFAR10-to-CIFAR10C, gradually changing | average error (%): 13.7 -> 10.4 | Helped |
| Use CoTTA instead of TENT-continual | WideResNet-28 · CIFAR10-to-CIFAR10C, gradually changing | average error (%): 30.7 -> 10.4 | Helped |
| Use CoTTA instead of BN Stats Adapt | ResNeXt-29 · CIFAR100-to-CIFAR100C | mean error rate (%): 35.4 -> 32.5 | Helped |
| Use CoTTA instead of TENT | ResNet-50 · ImageNet-to-ImageNet-C | average error (%): 66.5 -> 63.0 | Helped |
| Use CoTTA instead of the source-only model | Segformer-B5 · Cityscapes-to-ACDC | mean mIoU (%): 56.7 -> 58.6 | Helped |
| Add weight-averaged teacher predictions (ablation) | CoTTA · CIFAR10-to-CIFAR10C, standard | mean error rate (%): 20.7 -> 18.3 | Helped |
| Add augmentation-averaged predictions (ablation) | CoTTA with weight-averaged predictions · CIFAR10-to-CIFAR10C, standard | mean error rate (%): 18.3 -> 17.4 | Helped |
| Add stochastic restoration of weights to source weights (ablation) | CoTTA with weight and augmentation averaging · CIFAR10-to-CIFAR10C, standard | mean error rate (%): 17.4 -> 16.2 | Helped |
Closest papers by meaning
- A Versatile Framework for Continual Test-Time Domain Adaptation: Balancing Discriminability and Generalizability (CVPR 2024 · similarity 0.92)
- ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation (ICLR 2024 · similarity 0.90)
- DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation (ICML 2025 · similarity 0.90)
- PALM: Pushing Adaptive Learning Rate Mechanisms for Continual Test-Time Adaptation (AAAI 2025 · similarity 0.89)
- ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains (NeurIPS 2025 · similarity 0.89)
- Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization (AAAI 2024 · 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.