Research map / Probabilistic and scientific modeling / Causal discovery
Recovering Latent Causal Factor for Generalization to Distributional Shifts
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 |
|---|---|---|---|
| separate causal factor S from non-causal Z (LaCIM vs sVAE) (ablation) | LaCIM · CMNIST | accuracy (%): 93.6 -> 96.6 | Helped |
| separate causal factor S from non-causal Z (LaCIM vs sVAE) (ablation) | LaCIM · NICO m=14 | accuracy (%): 64.3 -> 66.4 | Helped |
| separate causal factor S from non-causal Z (LaCIM vs sVAE) (ablation) | LaCIM · ADNI D: Age | accuracy (%): 62.7 -> 63.8 | No clear effect |
| increase number of training domains from 8 to 14 (total samples fixed) | LaCIM · NICO | accuracy (%): 63.2 -> 66.4 | Helped |
| train on multiple domains instead of pooled data (LaCIM vs pool-LaCIM) (ablation) | LaCIM · simulation | mean correlation coefficient | Helped |
| increase domains from m=3 to m=5 in simulation (ablation) | LaCIM · simulation | mean correlation coefficient | Helped |
| use LaCIM instead of ERM | ERM · CMNIST | accuracy (%): 91.9 -> 96.6 | Helped |
| use LaCIM instead of IRM | IRM · NICO m=8 | accuracy (%): 61.4 -> 63.2 | Helped |
Closest papers by meaning
- Not all distributional shifts are equal: Fine-grained robust conformal inference (ICML 2024 · similarity 0.87)
- Identifying General Mechanism Shifts in Linear Causal Representations (NeurIPS 2024 · similarity 0.87)
- Causal Representation Learning from Multiple Distributions: A General Setting (ICML 2024 · similarity 0.86)
- Identification of Latent Confounders via Investigating the Tensor Ranks of the Nonlinear Observations (ICML 2025 · similarity 0.85)
- Quantifying Uncertainty in the Presence of Distribution Shifts (NeurIPS 2025 · similarity 0.85)
- Differentiable Causal Discovery for Latent Hierarchical Causal Models (ICLR 2025 · 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.