# When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning

arXiv: https://arxiv.org/abs/2510.07517 · HTML: https://mlsanity.com/paper/2510.07517/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 30 of its 74 references that are in the map.
- First posted on arXiv 8 Oct 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- 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, 30 of 30)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [Multi-Agent Collaboration Mechanisms: A Survey of LLMs](https://mlsanity.com/paper/2501.06322/) (2025)
- [Self-Preference Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2410.21819/) (2024)
- [Simple synthetic data reduces sycophancy in large language models](https://mlsanity.com/paper/2308.03958/) (2023)
- [Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems](https://mlsanity.com/paper/2410.02506/) (2024)
- [SycEval: Evaluating LLM Sycophancy](https://mlsanity.com/paper/2502.08177/) (2025)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [GroupDebate: Enhancing the Efficiency of Multi-Agent Debate Using Group Discussion](https://mlsanity.com/paper/2409.14051/) (2024)
- [Beyond Self-Talk: A Communication-Centric Survey of LLM-Based Multi-Agent Systems](https://mlsanity.com/paper/2502.14321/) (2025)
- [When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models](https://mlsanity.com/paper/2508.02087/) (2025)
- [Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity](https://mlsanity.com/paper/2502.08788/) (2025)
- [Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System](https://mlsanity.com/paper/2410.08115/) (2024)
- [Do LLM Evaluators Prefer Themselves for a Reason?](https://mlsanity.com/paper/2504.03846/) (2025)
- [Play Favorites: A Statistical Method to Measure Self-Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2508.06709/) (2025)
- [TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models](https://mlsanity.com/paper/2503.11656/) (2025)
- [Beyond the Surface: Measuring Self-Preference in LLM Judgments](https://mlsanity.com/paper/2506.02592/) (2025)
- [Persuade Me if You Can: A Framework for Evaluating Persuasion Effectiveness and Susceptibility Among Large Language Models](https://mlsanity.com/paper/2503.01829/) (2025)
- [Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics](https://mlsanity.com/paper/2506.12657/) (2025)
- [Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA](https://mlsanity.com/paper/2508.13743/) (2025)
- [Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework](https://mlsanity.com/paper/2408.11261/) (2024)
- [Exploring and Controlling Diversity in LLM-Agent Conversation](https://mlsanity.com/paper/2412.21102/) (2024)
- [When Persuasion Overrides Truth in Multi-Agent LLM Debates: Introducing a Confidence-Weighted Persuasion Override Rate (CW-POR)](https://mlsanity.com/paper/2504.00374/) (2025)
- [Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs](https://mlsanity.com/paper/2506.21561/) (2025)
- [Town Hall Debate Prompting: Enhancing Logical Reasoning in LLMs through Multi-Persona Interaction](https://mlsanity.com/paper/2502.15725/) (2025)
- [Silencer: From Discovery to Mitigation of Self-Bias in LLM-as-Benchmark-Generator](https://mlsanity.com/paper/2505.20738/) (2025)

## Built on by (most cited first, 21 of 21)

- [Not All Flips Are Conformity: Decomposing Stance Convergence in Multi-Agent LLM Debate](https://mlsanity.com/paper/2606.00820/) (2026)
- [Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems](https://mlsanity.com/paper/2604.02668/) (2026)
- [ModeX: Evaluator-Free Best-of-N Selection for Open-Ended Generation](https://mlsanity.com/paper/2601.02535/) (2026)
- [Epistemic Gain, Aleatoric Cost: Uncertainty Decomposition in Multi-Agent Debate for Math Reasoning](https://mlsanity.com/paper/2603.01221/) (2026)
- [Cross-Context Review: Improving LLM Output Quality by Separating Production and Review Sessions](https://mlsanity.com/paper/2603.12123/) (2026)
- [Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy](https://mlsanity.com/paper/2605.12991/) (2026)
- [The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size](https://mlsanity.com/paper/2606.02646/) (2026)
- [Emergence of Biased Consensus in Multi-Agent LLM Debates](https://mlsanity.com/paper/2608.02827/) (2026)
- [The Reasoning Trap: An Information-Theoretic Bound on Closed-System Multi-Step LLM Reasoning](https://mlsanity.com/paper/2605.01704/) (2026)
- [Self-Preference Is Weak or Absent in Verifiable Instruction-Following Revision: A Four-Model Test Under Genuine Authorship](https://mlsanity.com/paper/2606.20093/) (2026)
- [Peer Identity Bias in Multi-Agent LLM Evaluation: An Empirical Study Using the TRUST Democratic Discourse Analysis Pipeline](https://mlsanity.com/paper/2604.22971/) (2026)
- [Can Multi-Agent LLMs Identify Their Peers? Stylometric Fingerprinting in Role-Constrained Political Analysis](https://mlsanity.com/paper/2606.09854/) (2026)
- [Epistemic Blinding: An Inference-Time Protocol for Auditing Prior Contamination in LLM-Assisted Analysis](https://mlsanity.com/paper/2604.06013/) (2026)
- [When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaning](https://mlsanity.com/paper/2606.02866/) (2026)
- [Durable Evaluation Framework: Adversarial Arbitration for Sycophancy Reduction in Large Language Models](https://mlsanity.com/paper/2606.07532/) (2026)
- [Where Reasoning Diverges: Localized Multi-Agent Debate for Multi-Hop Question Answering](https://mlsanity.com/paper/2608.01463/) (2026)
- [The Civilization Framework: Sovereign-Anchored Communication Between Personal Multi-Agent Systems](https://mlsanity.com/paper/2609.03425/) (2026)
- [Trust the Brand, Lose Control: How Identity Hijacks LLM Agent Orchestration](https://mlsanity.com/paper/2609.32635/) (2026)
- [Prompted Identity Degrades Cooperation in Multi-Agent LLM Systems](https://mlsanity.com/paper/2609.35928/) (2026)
- [Beyond Final Accuracy: Auditing Communication in LLM Multi-Agent Systems](https://mlsanity.com/paper/2610.01042/) (2026)
- [Review Rarely Beats the Stronger Model: Separating Verification from Capability Transfer in LLM Pipelines](https://mlsanity.com/paper/2610.15897/) (2026)

## Abstract (arXiv)

Multi-agent debate (MAD) aims to improve large language model (LLM) reasoning by letting multiple agents exchange answers and then aggregate their opinions. Yet recent studies reveal that agents are not neutral: they are prone to identity-driven sycophancy and self-bias, uncritically adopting a peer's view or stubbornly adhering to their own prior output, undermining the reliability of debate. In this work, we present the first principled framework that joins sycophancy and self-bias to mitigate and quantify identity bias in MAD. First, we formalize the debate dynamics as an identity-weighted Bayesian update process. Second, we propose response anonymization: by removing identity markers from prompts, agents cannot distinguish "self" from "peer", which forces equal weights on agent identity, thereby reducing bias and improving trustworthiness. Third, we define the Identity Bias Coefficient (IBC), a principled bias metric that measures an agent's tendency to follow its peer versus itself. Empirical studies across multiple models and benchmarks confirm that identity bias is widespread, with sycophancy far more common than self-bias. Our findings highlight the need to ensure that MAD systems reason based on content rather than identity. Code is released in https://github.com/deeplearning-wisc/MAD-identity-bias.