[1F3-OS-10a] OS-10

1:55 PM - 2:10 PM JST(4:55 AM - 5:10 AM UTC)

[1F3-OS-10a-02] The Impact of Protocol Mechanics and Model Capacity on Emergent Behavior in Multilateral LLM Negotiations

〇Cifong Kang¹, Takehisa Yairi¹ (1. The University of Tokyo)

Keywords:

Multi-Agent Systems,Decision Making & Consensus Building,Game Theory

The integration of Large Language Models into Multi-Agent Systems challenges traditional game-theoretic predictions, particularly regarding the user dynamics under different voting rules. This study investigates negotiation protocols as governing mechanisms for LLM behavior, comparing high-capacity and lightweight counterparts under varying voting rules and turn-taking structures. Our results reveal a sharp divergence: lightweight models exhibit premature convergence, rushing into low-quality decisions. While rigid turn-taking exacerbates a last-mover disadvantage for these agents, procedural stochasticity acts as a cognitive prosthetic, mitigating positional biases. Conversely, high-capacity models display alignment resilience, refusing to exploit majority rules to marginalize minorities. Instead, they consistently converge near Pareto, Nash, and Kalai-Smorodinsky optima. We attribute this to a conflict between protocol rationality and RLHF alignment biases. These findings underscore that future mechanism design must be capacity-aware, accounting for the variable cognitive constraints of AI agents.