Research map / Label-efficient and robust learning / OOD and anomaly detection
ZooD: Exploiting Model Zoo for Out-of-Distribution Generalization
Qishi Dong, Muhammad Awais, Fengwei Zhou, Chuanlong Xie, Tianyang Hu, Yongxin Yang, Sung-Ho Bae, Zhenguo 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 the top-1 ranked model with a linear head instead of ERM | ZooD Single · Office-Home | accuracy: 66.5 -> 84.6 | Improved |
| use the top-1 ranked model instead of SWAD | ZooD Single · PACS | accuracy: 88.1 -> 96.0 | Improved |
| use the top-1 ranked model instead of SWAD | ZooD Single · TerraIncognita | accuracy: 50.0 -> 37.3 | Hurt |
| fine-tune the whole top-ranked model instead of only the classifier | ZooD Single · TerraIncognita | accuracy: +2.6% | Improved |
| full fine-tuning instead of top-classifier fine-tuning (ablation) | ZooD top model · PACS | accuracy: 90.6 -> 96.0 | Hurt |
| full fine-tuning instead of top-classifier fine-tuning (ablation) | ZooD top model · Office-Home | accuracy: 83.4 -> 84.6 | Hurt |
| ensemble top-K models without feature selection vs single model (ablation) | ZooD Ensemble · DomainNet | accuracy: 48.2 -> 50.5 | Improved |
| ensemble top-K models without feature selection vs single model (ablation) | ZooD Ensemble · TerraIncognita | accuracy: 37.3 -> 38.2 | Mixed |
| add variational EM feature selection to the ensemble (ablation) | ZooD · DomainNet | accuracy: 50.5 -> 50.6 | Improved |
| add feature selection to the ensemble (ablation) | ZooD · TerraIncognita | accuracy: 38.2 -> 42.3 | Improved |
| add feature selection to the ensemble (ablation) | ZooD · Office-Home | accuracy: 85.0 -> 85.1 | No clear effect |
| vary number of ensembled models K in {3, 5, 7} (ablation) | ZooD · OoD benchmarks | accuracy: K=3 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SWAD — SWAD: Domain Generalization by Seeking Flat Minima (NeurIPS 2021) · better on 5 of 6 shared results
- LogME — LogME: Practical Assessment of Pre-trained Models for Transfer Learning (ICML 2021) · 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
- OOD-Chameleon: Is Algorithm Selection for OOD Generalization Learnable? (ICML 2025 · similarity 0.88)
- CRoFT: Robust Fine-Tuning with Concurrent Optimization for OOD Generalization and Open-Set OOD Detection (ICML 2024 · similarity 0.87)
- Unraveling the Key Components of OOD Generalization via Diversification (ICLR 2024 · similarity 0.87)
- Spurious Feature Diversification Improves Out-of-distribution Generalization (ICLR 2024 · similarity 0.87)
- LCA-on-the-Line: Benchmarking Out of Distribution Generalization with Class Taxonomies (ICML 2024 · similarity 0.87)
- AHA: Human-Assisted Out-of-Distribution Generalization and Detection (NeurIPS 2024 · 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.
