spurious-correlations

[ICML 2023] Change is Hard: A Closer Look at Subpopulation Shift

Implementation of the EMNLP 2020 paper "Counterfactual Generator: A Weakly-Supervised Method for Named Entity Recognition".

Spurious Features Everywhere - Large-Scale Detection of Harmful Spurious Features in ImageNet

[Nature Medicine] The Limits of Fair Medical Imaging AI In Real-World Generalization

[NeurIPS 2023] Understanding and Improving Feature Learning for Out-of-Distribution Generalization

Official Repository for Spurious Memorization - Nature Communications 2025

Shortcut to Nowhere: Demystifying Deep Spurious Regression

[ICCV25] TD-DETR: The Devil is in the Spurious Correlations: Boosting Moment Retrieval with Dynamic Learning

Carloni, G., Tsaftaris, S. A., & Colantonio, S. (2024). CROCODILE: Causality aids RObustness via COntrastive DIsentangled LEarning @ MICCAI 2024 UNSURE Workshop

The official repository for "Training Debiased Subnetworks with Contrastive Weight Pruning"

This repository contains the code for the paper "Disentangling representations of retinal images with generative models".

Codebase for "Aggregation Hides Out-of-Distribution Generalization Failures from Spurious Correlations" (NeurIPS 25 Spotlight)

Mitigating Spurious Correlations for Self-supervised Recommendation

Diagnosing shortcut learning and spurious correlations in Waterbirds with group-aware evaluation, group-balanced training, and Grad-CAM.

UNCP (Universal Noise Calibration Protocol) — a domain-agnostic framework for mitigating spurious correlations in deep learning via targeted noise injection. No group labels required.

Solution for the Third Challenge of the First Phase in the Rayan International AI Contest: Hierarchical Group Robustness.

Reproduction code + artifacts for “When Do Deep Ensembles Improve Robustness to Spurious Correlations?” (TU Delft, 2026).

Making neural networks OOD robust using dissimilarity loss

This is the official implementation of our paper Robust Hate Speech Detection via Mitigating Spurious Correlations (Tiwari et al., AACL-IJCNLP 2022)