Debiasing Without Annotations: Cross-View Consensus Boosts Worst-Group Accuracy
Weiwei Li · hf · 2026-11-19
Researchers from UESTC and Sichuan University propose a multi-view debiasing method that aggregates within-class partitions across diverse representation spaces to reduce reliance on spurious correlations. Unlike prior approaches requiring costly human-annotated bias attributes—which may not match the shortcuts models actually use—this cross-view consensus approach improves worst-group accuracy without any bias supervision.
More from Research
- Tsinghua-led ALIVE makes streaming video models interactive with proactive monitoring and feedback-seeking — Tsinghua · 2026-11-19
- Stanford removes the decoder: direct pixel prediction from diffusion transformers cuts cost, lifts quality — StanfordUniversity · 2026-11-19
- AnisoFlow learns anisotropic prediction targets for flow models, boosting FID without touching architecture — StanfordUniversity · 2026-11-19
- KAIST Proposes PROLA: Low-Rank Adaptation for Spatiotemporal Weather Downscaling — MLVLab · 2026-11-19
- NTU's PACT Adapts Time-Series Foundation Models Without Forgetting Old Conditions — NanyangTechnologicalUniversity · 2026-11-19
- The Resolution Hypothesis: agent evolution stalls because noisy evals can't spot real improvements — ShanghaiTech · 2026-11-19