mlsanityData to 14 Oct 2026

Emergence of Biased Consensus in Multi-Agent LLM Debates

arXiv 2608.02827 ↗First posted 3 Aug 2026cs.MAcited by 3

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Multi-agent LLM debates achieve strong performance on decision-making tasks as well as problem-solving benchmarks, yet their safety and fairness risks remain poorly understood. Notably, interaction can amplify the biases of single LLMs, raising concerns for real-world deployment. We identify the emergence of collective (often biased) norms in multi-agent LLM debates and show that noise (e.g., LLM sampling temperature) is a key driver. To explain this, we propose an analytical framework drawing on physics-inspired theoretical models of social dynamics. We predict a phase transition to collective bias when conformity surpasses a critical threshold given the LLMs' initial bias and debate noise. We test the theoretical predictions through controlled experiments and observe a finite-size crossover consistent with an underlying phase transition. We further find that agent heterogeneity suppresses emergence by smoothing (rounding) this transition. Finally, we show that these insights generalize to realistic decision-making tasks, including investment decisions and LLM-as-a-judge evaluation.

Key facts. Cited by 3 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 44 of its 73 references that are in the map. First posted on arXiv 3 Aug 2026.

Builds on

44 of 73 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning8 Oct
  2. 2025Emergent Social Dynamics of LLM Agents in the El Farol Bar Problem4 Sep
  3. 2025Your AI Bosses Are Still Prejudiced: The Emergence of Stereotypes in LLM-Based Multi-Agent Systems27 Aug
  4. 2025Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?24 Aug
  5. 2025An Empirical Study of Group Conformity in Multi-Agent Systems2 Jun
  6. 2025Multiple LLM Agents Debate for Equitable Cultural Alignment30 May
  7. 2025From Belief Entrenchment to Robust Reasoning in LLM Agents21 Mar
  8. 2025Voting or Consensus? Decision-Making in Multi-Agent Debate26 Feb
  9. 2025Emergence of polarization in networks of large language model agents9 Jan
  10. 2024Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions3 Oct
  11. 2023Simulating Opinion Dynamics with Networks of LLM-based Agents16 Nov
  12. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate30 May

Other references · 32

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  3. 2023Large Language Models are not Fair Evaluatorscited by 1,085
  4. 2023Whose Opinions Do Language Models Reflect?cited by 773
  5. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviorscited by 598
  6. 2024Justice or Prejudice? Quantifying Biases in LLM-as-a-Judgecited by 368
  7. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agentscited by 367
  8. 2025Persona Vectors: Monitoring and Controlling Character Traits in Language Modelscited by 345
  9. 2023ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMscited by 309
  10. 2023Evaluating Large Language Models at Evaluating Instruction Followingcited by 302
  11. 2023Benchmarking Cognitive Biases in Large Language Models as Evaluatorscited by 150
  12. 2023Cultural Bias and Cultural Alignment of Large Language Modelscited by 149
20 more
  1. 2023Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language Modelscited by 148
  2. 2024Emergent social conventions and collective bias in LLM populationscited by 123
  3. 2024FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Makingcited by 113
  4. 2024Improving Multi-Agent Debate with Sparse Communication Topologycited by 107
  5. 2024Large Language Models are Geographically Biasedcited by 93
  6. 2024The Political Preferences of LLMscited by 84
  7. 2024Large Language Models Reflect the Ideology of their Creatorscited by 57
  8. 2024The Silicon Ceiling: Auditing GPT's Race and Gender Biases in Hiringcited by 29
  9. 2025The Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake Newscited by 29
  10. 2025Debate Only When Necessary: Adaptive Multiagent Collaboration for Efficient LLM Reasoningcited by 27
  11. 2024JobFair: A Framework for Benchmarking Gender Hiring Bias in Large Language Modelscited by 21
  12. 2024Mitigating the Risk of Health Inequity Exacerbated by Large Language Modelscited by 21
  13. 2024Teaching Models to Balance Resisting and Accepting Persuasioncited by 19
  14. 2015The Spontaneous Emergence of Conventions: An Experimental Study of Cultural Evolutioncited by 17
  15. 2023Confronting LLMs with Traditional ML: Rethinking the Fairness of Large Language Models in Tabular Classificationscited by 15
  16. 2025FairEval: Evaluating Fairness in LLM-Based Recommendations with Personality Awarenesscited by 8
  17. 2025Unequal Opportunities: Examining the Bias in Geographical Recommendations by Large Language Modelscited by 7
  18. 2025Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Modelscited by 6
  19. 2024Standard Language Ideology in AI-Generated Languagecited by 5
  20. 2025AccessEval: Benchmarking Disability Bias in Large Language Modelscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agentscited by 6
  2. 2026Xeno-Interpretability: Investigating the Alien Minds of LLMscited by 0
  3. 2026Multi-agent discussion gains less when dissent is withheldcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Group size effects and collective misalignment in LLM multi-agent systems
  2. 2025The Social Cost of Intelligence: Emergence, Propagation, and Amplification of Stereotypical Bias in Multi-Agent Systems
  3. 2026The Cost of Consensus: Isolated Self-Correction Prevails Over Unguided Homogeneous Multi-Agent Debate
  4. 2026Conformity Dynamics in LLM Multi-Agent Systems: The Roles of Topology and Self-Social Weighting
  5. 2026Multi-Agent Consensus as a Cognitive Bias Trigger in Human-AI Interaction
  6. 2025Can LLM Agents Really Debate? A Controlled Study of Multi-Agent Debate in Logical Reasoning
  7. 2025The Hunger Game Debate: On the Emergence of Over-Competition in Multi-Agent Systems
  8. 2026Sequential Consensus for Multi-Agent LLM Debates: A Wald-SPRT compute governor with calibration-based failure detection