# Prompted Identity Degrades Cooperation in Multi-Agent LLM Systems

arXiv: https://arxiv.org/abs/2609.35928 · HTML: https://mlsanity.com/paper/2609.35928/

## Key facts

- No citing papers in the map yet (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 61 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## 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)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [Mixture-of-Agents Enhances Large Language Model Capabilities](https://mlsanity.com/paper/2406.04692/) (2024)
- [Self-Preference Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2410.21819/) (2024)
- [Scaling Large Language Model-based Multi-Agent Collaboration](https://mlsanity.com/paper/2406.07155/) (2024)
- [MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents](https://mlsanity.com/paper/2503.01935/) (2025)
- [Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement](https://mlsanity.com/paper/2402.11436/) (2024)
- [Generative Language Models Exhibit Social Identity Biases](https://mlsanity.com/paper/2310.15819/) (2023)
- [Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation](https://mlsanity.com/paper/2309.17234/) (2023)
- [Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence](https://mlsanity.com/paper/2407.07061/) (2024)
- [MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning](https://mlsanity.com/paper/2502.18439/) (2025)
- [LLM-Coordination: Evaluating and Analyzing Multi-agent Coordination Abilities in Large Language Models](https://mlsanity.com/paper/2310.03903/) (2023)
- [BattleAgentBench: A Benchmark for Evaluating Cooperation and Competition Capabilities of Language Models in Multi-Agent Systems](https://mlsanity.com/paper/2408.15971/) (2024)
- [When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning](https://mlsanity.com/paper/2510.07517/) (2025)
- [Autonomous Agents for Collaborative Task under Information Asymmetry](https://mlsanity.com/paper/2406.14928/) (2024)
- [When Agents See Humans as the Outgroup: Belief-Dependent Bias in LLM-Powered Agents](https://mlsanity.com/paper/2601.00240/) (2026)
- [Evaluating Cooperation in LLM Social Groups through Elected Leadership](https://mlsanity.com/paper/2604.11721/) (2026)
- [Quantifying Self-Preservation Bias in Large Language Models](https://mlsanity.com/paper/2604.02174/) (2026)
- [Cooperative Profiles Predict Multi-Agent LLM Team Performance in AI for Science Workflows](https://mlsanity.com/paper/2604.20658/) (2026)

## Abstract (arXiv)

Multi-agent LLM systems increasingly mix models from several providers, yet exposing each agent's underlying model identity to its peers significantly impairs cooperation. We show that when agents are aware of each other's model family, the group splits into clusters, where agents prefer interacting with others carrying their same label, although nothing in the task rewards or asks for such a split. We argue that the label itself causes this split, which we define as textitfactionalism. We show and measure this phenomenon in two cooperative games and on a reasoning benchmark, with nine to twenty-five agents drawn from up to five open-weight model families. We further show that when the announced families are shuffled, or replaced by arbitrary labels, the factions still follow this information; when the label is removed, this behavior disappears. In strictly cooperative tasks, labeled groups spend on average 30% more rounds and 55% more tokens to reach a decision, and their success rate drops from 96% to 81%. The effect replicates across tasks, group sizes and model families. Withholding identity labels from the agents is simple and effective mitigation.