Latest Articles for Bias

A new method improves model fairness and accuracy without extensive group data.

A new method enhances fairness in colorectal cancer polyp detection.

Exploring how LLMs can assess model outputs in multiple languages.

Counterfactual explanations clarify AI predictions by showing possible input changes.

This study examines how patient self-diagnosis affects language model accuracy in medicine.

Study shows AI tools can improve accuracy in clinical decisions.

Analyzing gender bias in language models through Natural Language Inference.

This research addresses bias in document retrieval systems using uncertainty estimates.

Examining biases in how-to guides based on audience.

Enhancing Zero-Shot NAS using bias correction for better model performance.

A method for fair machine learning that reduces bias in decision-making systems.

Examining how biases affect the quality of language model evaluations.

A new framework identifies and measures bias in image generation systems.

New framework SUDO helps assess AI systems without ground-truth data.

This paper presents a method to improve fairness in federated learning while safeguarding data privacy.

A study evaluates various fairness improvement techniques for deep learning models in image classification.

New skin tone metrics aim to improve fairness and privacy in face recognition systems.

Examining the trade-offs in human-AI interactions and their societal impacts.

A look at how LLMs enhance NLG evaluation methods.

Examining fairness and accuracy in detecting vulnerable road users by autonomous vehicles.

Learn how FAIR principles guide ethical data management for AI models.

Examining the impact of sentiment analysis on Bengali identities and biases.

A look at algorithm auditing research and its societal impacts.

Examining how LLMs prefer generated contexts over retrieved ones in answering questions.

Automated tools often misjudge African American English, leading to unfair treatment online.

Conference discusses fairness across image upsampling techniques and racial representation.

This study analyzes character objectification in movies using a novel AI approach.

A structured approach to understanding decision-making and biases.

A new framework aims to ensure equity in GNN predictions.

Study reveals bias against disabled candidates in AI hiring systems.

Examining biases in predictive models for real estate pricing using geospatial data.

Examining how summarization models reflect bias in political opinions.

A study examines the performance and biases of AI language models globally.

A new dataset and method enhance language model question generation.

Survey reveals mixed opinions on AI's efficiency and fairness in transportation.

This article discusses using aggregate data in machine learning while protecting privacy.

A method for verifying machine learning models to enhance trust and transparency.

A new approach enhances smartphone GNSS positioning accuracy by correcting pseudorange biases.

This article investigates gender and racial biases in large vision-language models.

A review of randomized controlled trials on hydroxychloroquine for Covid-19.