Research map / Label-efficient and robust learning / Recognition and long-tail
Fairness via Representation Neutralization
Mengnan Du, Subhabrata Mukherjee, Guanchu Wang, Ruixiang Tang, Ahmed Hassan Awadallah, Xia Hu
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 |
|---|---|---|---|
| Use GCE instead of standard cross entropy to train the bias-amplified model (ablation) | MEPS proxy annotation generation · MEPS | DP: 0.866 -> 0.839 | Hurt |
| Replace GCE proxy labels with cross-entropy-generated proxy labels (ablation) | RNF · MEPS | DP: 0.964 -> 0.936 | Hurt |
| Replace GCE proxy labels with random sensitive-attribute annotations (ablation) | RNF · MEPS | DP: 0.964 -> 0.902 | Hurt |
| Debias only the last layer instead of the last several layers (ablation) | RNF · MEPS | DP: 0.964 -> 0.971 | Improved |
| Debias only the last layer instead of the last several layers (ablation) | RNF · CelebA | DP: 0.836 -> 0.966 | Improved |
| Apply RNF on top of an EOR-debiased encoder (ablation) | RNF · MEPS | DP: 0.964 -> 0.980 | Improved |
| Apply RNF on top of an adversarially debiased encoder (ablation) | RNF · MEPS | DP: 0.964 -> 0.971 | Improved |
| Use ground-truth sensitive-attribute annotations instead of proxy annotations (ablation) | RNF · all datasets | fairness mitigation: 10% | Improved |
| Neutralize representations of same-label samples from different sensitive groups and retrain the classification head | RNF · Adult | DP-Similarity: 0.272 -> 0.075 | Improved |
Closest papers by meaning
- FairNet: Dynamic Fairness Correction without Performance Loss via Contrastive Conditional LoRA (NeurIPS 2025 · similarity 0.89)
- FairSIN: Achieving Fairness in Graph Neural Networks through Sensitive Information Neutralization (AAAI 2024 · similarity 0.88)
- Rethinking Fair Representation Learning for Performance-Sensitive Tasks (ICLR 2025 · similarity 0.86)
- Adversarial Fairness Network (AAAI 2024 · similarity 0.86)
- Classes Are Not Equal: An Empirical Study on Image Recognition Fairness (CVPR 2024 · similarity 0.86)
- Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference (NeurIPS 2025 · similarity 0.86)
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.
