Thursday, August 21, 2025
Science
No Result
View All Result
  • Login
  • HOME
  • SCIENCE NEWS
  • CONTACT US
  • HOME
  • SCIENCE NEWS
  • CONTACT US
No Result
View All Result
Scienmag
No Result
View All Result
Home Science News Technology and Engineering

Zhu & Lee to study data-centric social bias mitigation

June 3, 2024
in Technology and Engineering
Reading Time: 2 mins read
0
65
SHARES
590
VIEWS
Share on FacebookShare on Twitter
ADVERTISEMENT

Ziwei Zhu, Assistant Professor, Computer Science, and Jin Lee, Assistant Professor, Criminology, Law and Society, are set to receive funding for: “Data-Centric Social Bias Mitigation for Large Language Model-based Cyberharassment Detection.” 

Ziwei Zhu, Assistant Professor, Computer Science, and Jin Lee, Assistant Professor, Criminology, Law and Society, are set to receive funding for: “Data-Centric Social Bias Mitigation for Large Language Model-based Cyberharassment Detection.” 

ADVERTISEMENT

The researchers aim to address the critical challenge of algorithmic bias in Large Language Models (LLMs) used for cyberharassment detection. 

They will focus on reducing unfair treatment against minority populations identified by gender, race, religion, sexual orientation, and disability. 

Despite the advanced capabilities of LLMs in text classification and reasoning, their effectiveness is compromised by inherent social biases, leading to the misclassification of benign content as harassment. 

The researchers will undertake a comprehensive, data-centric strategy comprising four main tasks: (1) Developing a large-scale and broad benchmark for examining social bias in detecting cyberharassment, covering multiple types of cyberharassment and bias dimensions; (2) Adapting the Contact Hypothesis from social psychology to augment prompts for debiasing LLMs in the zero-shot setting; (3) Selecting effective demonstration data via optimization to debias LLMs in the in-context learning setting; and (4) Creating an effective dataset integrating Contact Hypothesis and the BBQ dataset for debiasing LLMs through fine-tuning. 

Additionally, they will produce an open-source library consolidating all developed data and methodologies. Through these efforts, they aim to significantly improve the inclusion and accessibility of LLM applications in cybersecurity, contributing to a safer and more inclusive digital environment.

Zhu and Lee will receive $50,000 from the Virginia Innovation Partnership Authority for this project. Funding will begin in June 2024 and will end in late May 2025.

###

ABOUT GEORGE MASON UNIVERSITY

George Mason University is Virginia’s largest public research university. Located near Washington, D.C., Mason enrolls more than 40,000 students from 130 countries and all 50 states. Mason has grown rapidly over the past half-century and is recognized for its innovation and entrepreneurship, remarkable diversity, and commitment to accessibility. In 2023, the university launched Mason Now: Power the Possible, a one-billion-dollar comprehensive campaign to support student success, research, innovation, community, and stewardship. Learn more at gmu.edu.



Share26Tweet16
Previous Post

The coldest lab in New York has a new quantum offering

Next Post

Avocados are ‘bad’ and vegans are ridiculous: How we justify eating too much meat

Related Posts

Technology and Engineering

KRISS Pioneers Domestic Development of Core Radar Components in Stealth Technology

August 21, 2025
blank
Medicine

Ultrafast Plasma Membrane Ca2+ Transport Mechanism

August 21, 2025
blank
Technology and Engineering

Ultrahigh-Precision Plasmonic Meta-Rotary Wave Oscillator

August 21, 2025
blank
Technology and Engineering

Noncommutative Metasurfaces: Pioneering New Frontiers in Quantum Entanglement

August 21, 2025
blank
Medicine

Proximity Screening Boosts Graphene’s Electronic Quality

August 21, 2025
blank
Technology and Engineering

Revolutionary Laser Technique Simplifies Production of High-Performance Alloy Films

August 21, 2025
Next Post

Avocados are 'bad' and vegans are ridiculous: How we justify eating too much meat

  • Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study

    Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study

    27536 shares
    Share 11011 Tweet 6882
  • University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept

    951 shares
    Share 380 Tweet 238
  • Bee body mass, pathogens and local climate influence heat tolerance

    641 shares
    Share 256 Tweet 160
  • Researchers record first-ever images and data of a shark experiencing a boat strike

    508 shares
    Share 203 Tweet 127
  • Warm seawater speeding up melting of ‘Doomsday Glacier,’ scientists warn

    311 shares
    Share 124 Tweet 78
Science

Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!

RECENT NEWS

  • KRISS Pioneers Domestic Development of Core Radar Components in Stealth Technology
  • Hospitals and Sanitation Practices Drive Antibiotic Resistance Spread in Guatemala
  • Beta-Blockers Reduce Mortality in Cancer Patients
  • Ultrafast Plasma Membrane Ca2+ Transport Mechanism

Categories

  • Agriculture
  • Anthropology
  • Archaeology
  • Athmospheric
  • Biology
  • Bussines
  • Cancer
  • Chemistry
  • Climate
  • Earth Science
  • Marine
  • Mathematics
  • Medicine
  • Pediatry
  • Policy
  • Psychology & Psychiatry
  • Science Education
  • Social Science
  • Space
  • Technology and Engineering

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 4,859 other subscribers

© 2025 Scienmag - Science Magazine

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • HOME
  • SCIENCE NEWS
  • CONTACT US

© 2025 Scienmag - Science Magazine

Discover more from Science

Subscribe now to keep reading and get access to the full archive.

Continue reading