Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Multi-task adversarial learning detects intersectional algorithmic bias in AI recruitment systems.

Jiwei Wang¹, Yan Xu¹, Ruifeng Liu²

¹School of Management, Lanzhou Technology and Business College, No. 68, Weile Avenue, Heping Development Zone, Lanzhou, 730101, Gansu, China.

Scientific Reports

|May 20, 2026

Summary

This summary is machine-generated.

Related Concept Videos

Halo Effect

Halo Effect

The halo effect is a cognitive bias in which an individual's overall impression influences judgments about their specific traits. This psychological phenomenon leads people to associate positive characteristics with those they perceive as generally good and negative characteristics with those they view as bad. This effect is particularly influential in social perception, professional evaluations, and decision-making processes.The Psychological Basis of the Halo EffectThe halo effect is rooted...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

High-Precision MEMS Resonant Pressure Sensor for Real-Time Barometric Monitoring.

Micromachines·2026

Same author

DFT study of irradiation damage-defect correlations with mechanical properties in uranium nitride.

Scientific reports·2026

Same author

Chinese Cancer Patients' Willingness for Phase III Trials: Health Belief Model and Social Support.

Health education & behavior : the official publication of the Society for Public Health Education·2026

Same author

Perceived Social Support Mediating Caregiver Burden and Mental Health in Schizophrenia Caregivers.

Health education & behavior : the official publication of the Society for Public Health Education·2026

Same author

Temperature-Dependent Microstructure Evolution and Superplastic Deformation Behavior of Cold-Deformed Cr4Mo4Ni4V Martensitic Steel: From Continuous to Discontinuous Dynamic Recrystallization.

Materials (Basel, Switzerland)·2026

Same author

Multimodality imaging for diagnosing and monitoring immunoglobulin G4-related coronary arteritis presenting as giant aneurysm: a case description.

Quantitative imaging in medicine and surgery·2026

Same journal

Winter-associated downregulation of ovarian NR5A2 correlates with impaired follicle development in the striped hamster (Cricetulus barabensis).

Scientific reports·2026

Same journal

Both underestimation and overestimation of sleep duration predict mortality in older men with sleep disturbances.

Scientific reports·2026

Same journal

Predicting the flood susceptibility under land use and climate change scenarios using deep learning algorithms.

Scientific reports·2026

Same journal

Progress towards sustainable development and the urban-rural divide: an analysis of municipalities in Japan.

Scientific reports·2026

Same journal

Satellite-based analysis of precipitation across algeria's hydrographic watersheds (1983-2022) and their relationship with climate indices.

Scientific reports·2026

Same journal

Study on acoustic emission and infrared radiation characteristics of coal combination with different tectonic coal thickness.

Scientific reports·2026

See all related articles

This study introduces a deep learning framework to detect algorithmic bias in AI recruitment, enhancing fairness for job seekers. The novel approach improves bias detection accuracy and provides interpretable insights into fairness issues.

Area of Science:

Artificial Intelligence
Computer Science
Data Science

Background:

Algorithmic bias in AI recruitment systems poses significant threats to social fairness and job seeker rights.
Existing bias detection methods lack granularity, comprehensive fairness assessment, and interpretability.

Purpose of the Study:

To propose a deep learning-based framework for detecting algorithmic bias in AI recruitment.
To address limitations in detection granularity, fairness assessment, and interpretability of current methods.

Main Methods:

Developed a multi-task adversarial learning architecture with attention mechanisms for fine-grained intersectional bias detection.
Constructed a multi-dimensional evaluation system with nine metrics across group, individual, and causal fairness, using Pareto frontier analysis.

Keywords:

AI recruitment Algorithmic bias Fairness assessment Interpretability analysis Multi-task adversarial learning

Related Experiment Videos

Integrated an interpretability module utilizing SHAP values, Grad-CAM, and causal mediation analysis for bias source localization.

Main Results:

The proposed framework achieved a 12-18% point improvement in intersectional bias detection accuracy over traditional methods.
Demonstrated effective identification and localization of bias sources within AI recruitment systems.
Revealed fairness trade-off relationships through Pareto frontier analysis.

Conclusions:

The deep learning framework offers enhanced accuracy and interpretability for algorithmic bias detection in AI recruitment.
Provides robust technical support and theoretical guidance for improving fairness in AI-driven hiring processes.
Contributes to safeguarding social fairness and protecting job seekers' rights in the era of AI recruitment.