cs.CLNov 16, 2026

Corrupting Ties: Investigating Robustness to Bias Injection Across Multi-Agent Network Topologies

Authors: Cléa Chataigner, Mina Arzaghi, Florian Carichon, Pablo Piantanida, Golnoosh Farnadi

Organizations: MILA, Montréal, Québec, Canada · McGill University, Montréal, Québec, Canada · Paris-Saclay University, Paris, France · ILLS - International Laboratory on Learning Systems, Montréal, Québec, Canada · 

Abstract

AI systems are increasingly deployed in high-stakes decision-making contexts, where interactions among components can introduce fairness risks that remain undetected when auditing components or collective outcomes in isolation. As LLM-based multi-agent systems become more prevalent, understanding how biased information propagates through agent interactions is essential for assessing their fairness. We introduce a controlled bias injection framework that replaces selected agent responses with predefined biased ones, enabling an intervention-based analysis of how network topology and injection configurations shape system behavior. We propose decision divergence metrics that quantify the effects of these interventions at the agent, network, and system levels, and evaluate five topologies on three binary classification tasks using LLMs from the Qwen and Gemma families. Sequential networks show the greatest divergence, whereas fully connected and mesh networks show less. Injection position also matters: in star networks, injecting the central agent has larger effects than injecting peripheral agents. Models further differ in whether divergence follows the injected bias. These findings highlight the role of network structure in the propagation of biased information and motivate topology-aware auditing of fairness risks in multi-agent systems.

Figures & tables

Appendix figures & tables20 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Examining Agents' Bias Amplification versus Suppression in Multi-Agent Systems

    May 27, 2026Zejian Eric Wu, Zhongyi Jiang, Yuan Zhuang +1Multi-Agent LLM SystemsAlgorithmic Fairness

  2. Aligned Agents, Biased Swarm: Measuring Bias Amplification in Multi-Agent Systems

    Apr 10, 2026Keyu Li, Jin Gao, Dequan WangAlgorithmic BiasMulti-Agent Systems

  3. Bias Control and Injection for Multi-Agent Aggregation

    Oct 15, 2026Sungjae Lee, Soheil Abbasloo, Jabin Koo +1Multi-Agent ReasoningLLM Agent Reliability