Research map / Continual learning / Domain and prompt-based
Deja Vu: Continual Model Generalization for Unseen Domains
Chenxi Liu, Lixu Wang, Lingjuan Lyu, Chen Sun, Xiao Wang, Qi Zhu
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 RandMix augmentation (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average TDG, TDA, and FA accuracy | Improved |
| Add RandMix augmentation (ablation) | SHOT, GSFDA, and PCL · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Improved |
| Replace T2PL with Softmax pseudo-labeling (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
| Replace T2PL with SHOT pseudo-labeling (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
| Replace T2PL with BMD pseudo-labeling (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
| Add PCA loss (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Improved |
| Replace PCA with regular prototype learning (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
| Replace PCA with supervised contrastive learning (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
| Replace PCA with PCL (ablation) | RaTP · Digits (US→SD→SN→MM→MT) | average performance across TDG, TDA, and FA | Hurt |
Closest papers by meaning
- Complementary Domain Adaptation and Generalization for Unsupervised Continual Domain Shift Learning (ICCV 2023 · similarity 0.91)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.89)
- Enhancing Evolving Domain Generalization through Dynamic Latent Representations (AAAI 2024 · similarity 0.88)
- PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization (CVPR 2024 · similarity 0.87)
- DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation (ICML 2025 · similarity 0.87)
- Continuous Domain Generalization (NeurIPS 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.
