Research map / Label-efficient and robust learning / Domain adaptation and generalization
Geometric Rate–Distortion Invariance for Domain Generalization
Tong Liu, Sen Liang, Shuo Bai
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 geometric distortion term (ablation) | RDI · PACS | average accuracy: 86.4 -> 87.4 | Improved |
| add aggregated rate term (ablation) | RDI · PACS | average accuracy: 86.5 -> 87.4 | Improved |
| add geometric projection head alone to ERM (ablation) | ERM · PACS | average accuracy: 80.0 -> 82.6 | Improved |
| add rate-distortion objectives on top of projection head (ablation) | ERM+g · PACS | average accuracy: 82.6 -> 87.4 | Improved |
| add both RDI losses to ERM+g (ablation) | ERM+g · PACS Sketch | accuracy: 74.4 -> 82.3 | Improved |
| add projection head to baselines (ERM, CORAL, VREx, IB ERM) (ablation) | DomainBed baselines · DomainBed | accuracy: -2.16% to +0.69% | No clear effect |
| use RDI instead of ERM | ResNet-50 · DomainBed average | accuracy: 63.3 -> 65.7 | Improved |
| use RDI instead of GGA | ResNet-50 · PACS | accuracy: 86.4 -> 87.4 | Improved |
| use RDI instead of GGA | ResNet-50 · DomainNet | accuracy: 44.4 -> 44.2 | Hurt |
| use RDI instead of CORAL | ResNet-50 · VLCS | accuracy: 78.8 -> 78.2 | Hurt |
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.
- CORAL — CORAL: Disentangling Latent Representations in Long-Tailed Diffusion (NeurIPS 2025) · better on 5 of 6 shared results
- LFME — LFME: A Simple Framework for Learning from Multiple Experts in Domain Generalization (NeurIPS 2024) · 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
- Failure Modes of Domain Generalization Algorithms (CVPR 2022 · similarity 0.89)
- Robust Domain Generalization under Divergent Marginal and Conditional Distributions (NeurIPS 2026 · similarity 0.89)
- One-Step Generalization Ratio Guided Optimization for Domain Generalization (ICML 2025 · similarity 0.87)
- Unknown Domain Inconsistency Minimization for Domain Generalization (ICLR 2024 · similarity 0.87)
- Continuous Domain Generalization (NeurIPS 2025 · similarity 0.87)
- Rethinking Multi-domain Generalization with A General Learning Objective (CVPR 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.
