Latest Articles for Bias

A new method enhances data reliability in A/B testing.

Examining societal bias in Chinese language models and search engines.

Examining the principles and challenges of creating fair AI technologies.

A new approach to improve fairness and accuracy in facial recognition.

Examining the effects of updates on safety, bias, and authenticity in image generation.

A study on Graphair's effectiveness in fair graph representations and link prediction.

Examining the challenges of bias in voice responses and user perspectives.

Learn how to optimize effectively using biased information sources.

A new benchmark evaluates biases in language models used for medical diagnoses.

A new method reveals insights into prediction errors and model complexity.

This article discusses methods to reduce bias in text safety classifiers using ensemble models.

Examining how social factors influence machine learning outcomes in healthcare.

Exploring current gaps in data transparency practices across AI systems.

A new method aims to reduce bias in machine learning models for better fairness.

New approach improves regression estimates by effectively handling outliers linked to variables.

Examining fairness issues in transaction fraud detection systems.

Algorithms confront bias and delays in user preference feedback.

Examining the risks and challenges of AI technology in medical applications.

Examining key security issues related to large language models in AI.

This study examines the relationship between intrinsic and extrinsic bias metrics in NLP.

This article examines bias in TTI systems and the role of embeddings.

Examining how language processing tools influence the richness of communication.

This article addresses bias in graphical models and proposes a method for fair analysis.

Examining how biases affect digital content suggestions across demographics.

An analysis questions the necessity of major trials for streptokinase's effectiveness.

This study investigates sex bias in predicting pilot stress and fatigue.

HEARTS aims to improve stereotype detection in text while ensuring explainability and sustainability.

Research highlights the importance of fair diagnosis in respiratory illnesses.

This study analyzes the two-tower model's performance against bias in document ranking.

Investigating the link between role-play and biases in language models.

Researchers introduce the STOP dataset to analyze bias in language models.

This paper discusses fairness in selecting candidates for institutions amid biased evaluations.

This study investigates AI's role in salary negotiation advice and potential biases.

Research shows how algorithms shape user views on social networks.

This paper investigates fairness in blocking methods for entity matching systems.

Examining how different thinking styles affect survey responses and biases.

PeerArg enhances transparency and reliability in the peer review process.

FVNNs provide a fair approach to data analysis and predictions.

PASPO provides a fresh approach to complex allocation challenges across various fields.

This paper evaluates fairness issues in Retrieval-Augmented Generation methods.