Research map / Label-efficient and robust learning / OOD and anomaly detection
Learning Causal Semantic Representation for Out-of-Distribution Prediction
Read the paperCheck 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 CSG (separate semantic/variation latent factors) instead of CE | CE classifier · Shifted-MNIST δ0=δ1=0 | accuracy (%): 42.9 -> 81.4 | Helped |
| use independent prior for test-domain prediction (CSG-ind vs CSG) | CSG · Shifted-MNIST δ0=δ1=0 | accuracy (%): 81.4 -> 82.6 | Helped |
| use independent prior for test-domain prediction (CSG-ind vs CSG) | CSG · PACS others→S | accuracy (%): 80.7 -> 81.1 | Helped |
| adapt prior with unlabeled test data (CSG-DA vs CSG-ind) | CSG · Shifted-MNIST δ0=δ1=0 | accuracy (%): 82.6 -> 97.6 | Helped |
| apply domain adaptation methods (DANN, CDAN, MDD) to CE baseline | CE · Shifted-MNIST δ0=δ1=0 | accuracy (%): 42.9 -> 40.9 (DANN) | Hurt |
| use CSG instead of CE under pooled-domain training | CE · PACS others→C | accuracy (%): 77.9 -> 84.4 | Helped |
| use CSG instead of CE | CE · ImageCLEF-DA others→P / P→C | accuracy (%): 91.2 -> 92.3 (P→C) | Helped |
| use CSG instead of CE | CE · PACS others→P | accuracy (%): 97.8 -> 97.7 | No clear effect |
| CNBB causal discriminative baseline instead of CE | CE · PACS others→S | accuracy (%): 79.1 -> 43.3 | Hurt |
| model semantic s and variation v separately (vs treating them as a whole) (ablation) | CSG methods · Shifted-MNIST, ImageCLEF-DA, PACS | accuracy (%) | Helped |
Closest papers by meaning
- Toward Causal Generative Modeling: From Representation to Generation (AAAI 2025 · similarity 0.87)
- On the Out-of-Distribution Generalization of Self-Supervised Learning (ICML 2025 · similarity 0.86)
- Causality Inspired Federated Learning for OOD Generalization (ICML 2025 · similarity 0.85)
- Diffusion-based Layer-wise Semantic Reconstruction for Unsupervised Out-of-Distribution Detection (NeurIPS 2024 · similarity 0.84)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (NeurIPS 2024 · similarity 0.84)
- A Fixed-Point Approach for Causal Generative Modeling (ICML 2024 · similarity 0.84)
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.