spurious correlations
These are misleading statistical associations that arise from confounding variables or coincidence rather than a genuine causal relationship. In AI, especially in machine learning, spurious correlations can lead to models that generalize poorly to unseen data, prompting researchers to develop methods to identify and mitigate them.
- Aggregation Hides Out-of-Distribution Generalization Failures from Spurious Correlations
- Automated Detection of Visual Attribute Reliance with a Self-Reflective Agent
- Boosting Resilience of Large Language Models through Causality-Driven Robust Optimization
- Causality Meets the Table: Debiasing LLMs for Faithful TableQA via Front-Door Intervention
- Data Selection Matters: Towards Robust Instruction Tuning of Large Multimodal Models
- Dataset Distillation for Pre-Trained Self-Supervised Vision Models
- Diffusing DeBias: Synthetic Bias Amplification for Model Debiasing
- Dual Data Alignment Makes AI-Generated Image Detector Easier Generalizable
- Escaping the SpuriVerse: Can Large Vision-Language Models Generalize Beyond Seen Spurious Correlations?
- FedIGL: Federated Invariant Graph Learning for Non-IID Graphs
- Interpreting vision transformers via residual replacement model
- Language‑Bias‑Resilient Visual Question Answering via Adaptive Multi‑Margin Collaborative Debiasing
- Learning the Wrong Lessons: Syntactic-Domain Spurious Correlations in Language Models
- Learning to Focus: Causal Attention Distillation via Gradient‐Guided Token Pruning
- Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI
- Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention
- SHGR: A Generalized Maximal Correlation Coefficient
- Self-Supervised Discovery of Neural Circuits in Spatially Patterned Neural Responses with Graph Neural Networks
- Self-alignment of Large Video Language Models with Refined Regularized Preference Optimization
- Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
- The Impact of Coreset Selection on Spurious Correlations and Group Robustness
- The Rich and the Simple: On the Implicit Bias of Adam and SGD
- Towards Single-Source Domain Generalized Object Detection via Causal Visual Prompts