# Mitigating Social Bias in Large Language Models: A Multi-Objective Approach within a Multi-Agent Framework

arXiv: https://arxiv.org/abs/2412.15504 · HTML: https://mlsanity.com/paper/2412.15504/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 24 of its 41 references that are in the map.
- First posted on arXiv 20 Dec 2024.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 24 of 24)

- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate](https://mlsanity.com/paper/2308.07201/) (2023)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment](https://mlsanity.com/paper/2308.05374/) (2023)
- [Prompting GPT-3 To Be Reliable](https://mlsanity.com/paper/2210.09150/) (2022)
- [Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration](https://mlsanity.com/paper/2307.05300/) (2023)
- [The Capacity for Moral Self-Correction in Large Language Models](https://mlsanity.com/paper/2302.07459/) (2023)
- ["I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset](https://mlsanity.com/paper/2205.09209/) (2022)
- [Evaluating and Mitigating Discrimination in Language Model Decisions](https://mlsanity.com/paper/2312.03689/) (2023)
- [AI Transparency in the Age of LLMs: A Human-Centered Research Roadmap](https://mlsanity.com/paper/2306.01941/) (2023)
- [Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs](https://mlsanity.com/paper/2311.17371/) (2023)
- [Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey](https://mlsanity.com/paper/2210.07700/) (2022)
- [Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes](https://mlsanity.com/paper/2402.01981/) (2024)
- [Nationality Bias in Text Generation](https://mlsanity.com/paper/2302.02463/) (2023)
- [A Robust Bias Mitigation Procedure Based on the Stereotype Content Model](https://mlsanity.com/paper/2210.14552/) (2022)
- [Towards Understanding Gender Bias in Relation Extraction](https://mlsanity.com/paper/1911.03642/) (2019)
- [Reinforcement Learning from Multi-role Debates as Feedback for Bias Mitigation in LLMs](https://mlsanity.com/paper/2404.10160/) (2024)
- [Can Instruction Fine-Tuned Language Models Identify Social Bias through Prompting?](https://mlsanity.com/paper/2307.10472/) (2023)
- [Aligning Large Language Models with Counterfactual DPO](https://mlsanity.com/paper/2401.09566/) (2024)

## Built on by (most cited first, 8 of 8)

- [Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Prompts](https://mlsanity.com/paper/2602.04398/) (2026)
- [Breaking Language Barriers or Reinforcing Bias? A Study of Gender and Racial Disparities in Multilingual Contrastive Vision Language Models](https://mlsanity.com/paper/2505.14160/) (2025)
- [BiasFreeBench: a Benchmark for Mitigating Bias in Large Language Model Responses](https://mlsanity.com/paper/2510.00232/) (2025)
- [Bias Mitigation Agent: Optimizing Source Selection for Fair and Balanced Knowledge Retrieval](https://mlsanity.com/paper/2508.18724/) (2025)
- [Aligned Agents, Biased Swarm: Measuring Bias Amplification in Multi-Agent Systems](https://mlsanity.com/paper/2604.08963/) (2026)
- [ImplicitBBQ: Benchmarking Implicit Bias in Large Language Models through Characteristic Based Cues](https://mlsanity.com/paper/2604.01925/) (2026)
- [Fairness in Multi-Agent Systems for Software Engineering: An SDLC-Oriented Rapid Review](https://mlsanity.com/paper/2604.13103/) (2026)
- [Where are the Humans? A Scoping Review of Fairness in Multi-agent AI Systems](https://mlsanity.com/paper/2604.15078/) (2026)

## Abstract (arXiv)

Natural language processing (NLP) has seen remarkable advancements with the development of large language models (LLMs). Despite these advancements, LLMs often produce socially biased outputs. Recent studies have mainly addressed this problem by prompting LLMs to behave ethically, but this approach results in unacceptable performance degradation. In this paper, we propose a multi-objective approach within a multi-agent framework (MOMA) to mitigate social bias in LLMs without significantly compromising their performance. The key idea of MOMA involves deploying multiple agents to perform causal interventions on bias-related contents of the input questions, breaking the shortcut connection between these contents and the corresponding answers. Unlike traditional debiasing techniques leading to performance degradation, MOMA substantially reduces bias while maintaining accuracy in downstream tasks. Our experiments conducted on two datasets and two models demonstrate that MOMA reduces bias scores by up to 87.7%, with only a marginal performance degradation of up to 6.8% in the BBQ dataset. Additionally, it significantly enhances the multi-objective metric icat in the StereoSet dataset by up to 58.1%. Code will be made available at https://github.com/Cortantse/MOMA.