Bias Unveiled: Investigating Social Bias in LLM-Generated Code

2024/11/15 by Lin Ling, Fazle Rabbi, Ling, Lin +5 · 7 citations
Computer Science · Social Sciences · #Artificial Intelligence in Law #FOS: Computer and information sciences #Hate Speech and Cyberbullying Detection #Software Engineering (cs.SE)

paper · pdf · doi:10.48550/arxiv.2411.10351

openalex publication_date 2024/11/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) have significantly advanced the field of automated code generation. However, a notable research gap exists in evaluating social biases that may be present in the code produced by LLMs. To solve this issue, we propose a novel fairness framework, i.e., Solar, to assess and mitigate the social biases of LLM-generated code. Specifically, Solar can automatically generate test cases for quantitatively uncovering social biases of the auto-generated code by LLMs. To quantify the severity of social biases in generated code, we develop a dataset that covers a diverse set of social problems. We applied Solar and the crafted dataset to four state-of-the-art LLMs for code generation. Our evaluation reveals severe bias in the LLM-generated code from all the subject LLMs. Furthermore, we explore several prompting strategies for mitigating bias, including Chain-of-Thought (CoT) prompting, combining positive role-playing with CoT prompting and dialogue with Solar. Our experiments show that dialogue with Solar can effectively reduce social bias in LLM-generated code by up to 90%. Last, we make the code and data publicly available is highly extensible to evaluate new social problems.

Cited by

• AXIOM: Benchmarking LLM-as-a-Judge for Code via Rule-Based Perturbation and Multisource Quality Calibration
• A Multifaceted Analysis of Social Biases in Large Language Models
• BabelCoder: Agentic Code Translation with Specification Alignment
• CFCEval: Evaluating Security Aspects in Code Generated by Large Language Models
• What AI Speaks for Your Community: Polling AI Agents for Public Opinion on Data Center Projects
• The Bias is in the Details: An Assessment of Cognitive Bias in LLMs
• GPT and Prejudice: A Sparse Approach to Understanding Learned Representations in Large Language Models

Related

• FairCoder: Evaluating Social Bias of LLMs in Code Generation
• Bias Testing and Mitigation in LLM-based Code Generation
• Uncovering and Quantifying Social Biases in Code Generation
• BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs
• With a Grain of SALT: Are LLMs Fair Across Social Dimensions?