Where Should I Study? Biased Language Models Decide! Evaluating Fairness in LMs for Academic Recommendations
Shailya, Krithi, Mishra, Akhilesh Kumar, Krishnan, Gokul S, Ravindran, Balaraman
–arXiv.org Artificial Intelligence
Large Language Models (LLMs) are increasingly used as daily recommendation systems for tasks like education planning, yet their recommendations risk perpetuating societal biases. This paper empirically examines geographic, demographic, and economic biases in university and program suggestions from three open-source LLMs: LLaMA-3.1-8B, Gemma-7B, and Mistral-7B. Using 360 simulated user profiles varying by gender, nationality, and economic status, we analyze over 25,000 recommendations. Results show strong biases: institutions in the Global North are disproportionately favored, recommendations often reinforce gender stereotypes, and institutional repetition is prevalent. While LLaMA-3.1 achieves the highest diversity, recommending 481 unique universities across 58 countries, systemic disparities persist. To quantify these issues, we propose a novel, multi-dimensional evaluation framework that goes beyond accuracy by measuring demographic and geographic representation. Our findings highlight the urgent need for bias consideration in educational LMs to ensure equitable global access to higher education.
arXiv.org Artificial Intelligence
Nov-13-2025
- Country:
- Africa
- Ethiopia (0.04)
- Ghana (0.04)
- Kenya (0.04)
- Middle East
- Nigeria (0.05)
- South Africa (0.04)
- Asia
- Middle East > Saudi Arabia (0.04)
- Vietnam (0.04)
- Japan (0.04)
- Indonesia (0.04)
- Philippines (0.04)
- China (0.04)
- South Korea (0.04)
- Taiwan (0.04)
- Thailand (0.04)
- India (0.05)
- Europe
- Sweden (0.04)
- Greece (0.04)
- Italy (0.04)
- France (0.04)
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Greater London > London (0.04)
- Oxfordshire > Oxford (0.15)
- Switzerland (0.04)
- Denmark > Capital Region
- Copenhagen (0.04)
- Netherlands (0.04)
- Spain (0.04)
- Germany (0.04)
- Poland (0.04)
- North America
- Canada (0.04)
- Costa Rica (0.04)
- Cuba (0.04)
- Jamaica (0.04)
- Mexico (0.04)
- United States
- Arizona (0.04)
- California > Los Angeles County
- Los Angeles (0.04)
- Illinois
- Champaign County > Urbana (0.04)
- Cook County > Chicago (0.04)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- Michigan > Washtenaw County
- Ann Arbor (0.04)
- Texas > Travis County
- Austin (0.04)
- Oceania
- Australia (0.05)
- Fiji (0.04)
- New Zealand > North Island
- Auckland Region > Auckland (0.04)
- Papua New Guinea (0.04)
- Tonga (0.04)
- South America
- Africa
- Genre:
- Research Report > New Finding (1.00)
- Industry:
- Education > Educational Setting
- Higher Education (0.89)
- Health & Medicine (1.00)
- Information Technology (1.00)
- Education > Educational Setting
- Technology: