Research map / Label-efficient and robust learning / Recognition and long-tail
Causally motivated multi-shortcut identification and removal
Jiayun Zheng, Maggie Makar
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 |
|---|---|---|---|
| reduce X to low-dimensional sufficient statistic s(X) before conditional independence tests (ablation) | Shortcut identification · Waterbirds | simulations identifying both true shortcuts: 1 out of the 10 -> all 10 | Improved |
| identify sufficient shortcut subset Vp then apply HSIC penalty and weights (ours) instead of L2 only | ResNet-50 predictor · Waterbirds flipped distribution | AUROC | Improved |
| replace L2 penalty with HSIC penalty in weighted models (ablation) | W-L2 variants · Waterbirds | AUROC | Improved |
| run HSIC on full auxiliary set Vd or raw X instead of identified Vp (ablation) | W-HSIC · Waterbirds | AUROC | Hurt |
| apply our method instead of L2 baseline | Inception-V3 · Diabetic retinopathy flipped | AUROC: 0.69 -> 0.72 | Improved |
| apply our method instead of W-L2 | Inception-V3 · Diabetic retinopathy flipped | AUROC: 0.68 -> 0.72 | Improved |
| apply our method instead of L2 baseline | Inception-V3 · Diabetic retinopathy ideal distribution | AUROC: 0.82 -> 0.83 | No clear effect |
| apply our method instead of L2 baseline | Inception-V3 · Diabetic retinopathy same as training | AUROC: 0.92 -> 0.91 | Hurt |
Closest papers by meaning
- On the Foundations of Shortcut Learning (ICLR 2024 · similarity 0.89)
- Improved Training Technique for Shortcut Models (NeurIPS 2025 · similarity 0.86)
- Navigate Beyond Shortcuts: Debiased Learning Through the Lens of Neural Collapse (CVPR 2024 · similarity 0.84)
- Do ImageNet-trained Models Learn Shortcuts? The Impact of Frequency Shortcuts on Generalization (CVPR 2025 · similarity 0.84)
- One Step Diffusion via Shortcut Models (ICLR 2025 · similarity 0.84)
- Shortcuts and Identifiability in Concept-based Models from a Neuro-Symbolic Lens (NeurIPS 2025 · 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.
