Research map / Label-efficient and robust learning / Domain adaptation and generalization
SENTRY: Selective Entropy Optimization via Committee Consistency for Unsupervised Domain Adaptation
Viraj Prabhu, Shivam Khare, Deeksha Kartik, Judy Hoffman
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 consistency-based selective entropy minimization instead of minimizing entropy on all target instances (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 68.8 -> 79.5 | Improved |
| add consistency-based selection (instead of all instances) (ablation) | SENTRY · OfficeHome RS-UT Rw->Cl | per-class accuracy: 44.9 -> 56.8 | Improved |
| add entropy maximization on inconsistent instances (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 77.7 -> 79.5 | Improved |
| add entropy maximization on inconsistent instances (ablation) | SENTRY · OfficeHome RS-UT Rw->Cl | per-class accuracy: 55.3 -> 56.8 | Improved |
| add entropy maximization on incorrect instances (oracle selection) (ablation) | SENTRY oracle · DomainNet C->S | per-class accuracy: 84.3 -> 86.3 | Improved |
| add entropy maximization on incorrect instances (oracle selection) (ablation) | SENTRY oracle · OfficeHome RS-UT Rw->Cl | per-class accuracy: 77.7 -> 80.1 | Improved |
| use majority instead of unanimous voting (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 77.8 -> 79.5 | Improved |
| use majority instead of unanimous voting (ablation) | SENTRY · OfficeHome RS-UT Rw->Cl | per-class accuracy: 52.2 -> 56.8 | Improved |
| vary committee size k and RandAugment transforms N (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 76.8 to 79.5 | No clear effect |
| backpropagate on augmented rather than original target instances (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 73.1 -> 79.5 | Improved |
| add pseudo class-balanced target sampling (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 76.6 -> 78.6 | Improved |
| add source class-balanced sampling (ablation) | SENTRY · DomainNet C->S | per-class accuracy: 76.6 -> 77.1 | Improved |
Later papers that compared with this method
10 later accepted papers report SENTRY as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.
- SaTeen: Learning Structural Alignment for Continual Test-Time Adaptation (ICML 2026 · better on 2 of 2 shared results)
- CASUAL: Conditional Support Alignment for Domain Adaptation with Label Shift (AAAI 2025 · better on 1 of 1 shared result)
- Entropy is not Enough for Test-Time Adaptation: From the Perspective of Disentangled Factors (ICLR 2024 · better on 3 of 3 shared results)
- Domain Adaptation with Adversarial Training on Penultimate Activations (AAAI 2023 · better on 4 of 4 shared results)
- Make the U in UDA Matter: Invariant Consistency Learning for Unsupervised Domain Adaptation (NeurIPS 2023 · better on 3 of 4 shared results)
- Domain-Indexing Variational Bayes: Interpretable Domain Index for Domain Adaptation (ICLR 2023 · better on 3 of 3 shared results)
- Energy-based Self-Training and Normalization for Unsupervised Domain Adaptation (ICCV 2023 · better on 2 of 2 shared results)
- Adapting Self-Supervised Vision Transformers by Probing Attention-Conditioned Masking Consistency (NeurIPS 2022 · better on 5 of 5 shared results)
2 more not listed.
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
- Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation (ICLR 2018 · similarity 0.90)
- Make the U in UDA Matter: Invariant Consistency Learning for Unsupervised Domain Adaptation (NeurIPS 2023 · similarity 0.89)
- Guiding Pseudo-Labels With Uncertainty Estimation for Source-Free Unsupervised Domain Adaptation (CVPR 2023 · similarity 0.89)
- Prior Knowledge Guided Unsupervised Domain Adaptation (ECCV 2022 · similarity 0.89)
- Theoretical Performance Guarantees for Partial Domain Adaptation via Partial Optimal Transport (ICML 2025 · similarity 0.88)
- Pseudo-Calibration: Improving Predictive Uncertainty Estimation in Unsupervised Domain Adaptation (ICML 2024 · 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.
