Sharing the Road: LLM-Based Bargaining in Mixed-Autonomy Traffic

Large Language Models (LLMs) offer a flexible computational approach for modeling social interactions among heterogeneous agents. This paper develops a novel framework towards characterizing and understanding LLM-based population-level bargaining of shared road space between autonomous vehicle (AV) and human-driven vehicle (HV) agents in mixed-autonomy traffic. Through controlled experiments varying traffic density, agent objectives, and traffic-domain knowledge, we first characterize bargaining outcomes and observe heterogeneous agreement rates across different experimental conditions. To understand how the bargaining process shapes the outcomes, we introduce a behavioral taxonomy that jointly characterizes numerical bidding actions and communicative attitudes. We examine two complementary aspects of social interaction: inter-agent bargaining dynamics and private-to-public behavioral transformation. Results reveal distinct interaction patterns associated with agreement and breakdown, including greater constructive reciprocity in agreement trials and more frequent conflict escalation in breakdown trials. Furthermore, private and public communicative attitudes often differ even when numerical bids remain unchanged. These findings illustrate how LLM-based bargaining simulations can support a structured analysis of social interaction behaviors in mixed-autonomy traffic.

Large language model (LLM) agents are increasingly deployed in competitive multi-agent settings, raising fundamental questions about whether they converge to equilibria and how their strategic behavior can be characterized. In this paper, we study LLM agent interactions in two standard games: a network resource allocation game and a Cournot competition game. Rather than converging to Nash equilibria, we find that LLM agents tend to cooperate when given multi-round prompts and non-zero-sum context. Chain-of-thought analysis reveals that fairness reasoning is central to this behavior. We propose an analytical framework that captures the dynamics of LLM agent reasoning across rounds and explains these experimental findings.

Multi-vehicle cooperative autonomous driving enhances the safety and reliability of autonomous driving systems through information sharing among connected vehicles, demonstrating significant potential for improving traffic safety. LLM-based approaches leverage strong reasoning capabilities of LLMs to enable effective inter-vehicle negotiation and improve cooperative driving performance. However, driving decisions in complex traffic scenarios are inherently multi-solution in nature. As a result, existing negotiation-based methods often converge prematurely to suboptimal solutions, hindering consensus formation and limiting the practical deployment of cooperative autonomous driving systems. To address this challenge, we propose CoLMIN, the LLM-based multi-decision path negotiation framework for cooperative autonomous driving, achieving stable decision consensus through multi-decision path negotiation and reflective reasoning. To achieve stable and high-quality consensus in cooperative autonomous driving, CoLMIN consists of three key components: (i) an LLM-based Multi-Intent Negotiation module (LMin), which adopts a Negotiator-Evaluator paradigm and generates multiple candidate driving intentions for joint evaluation; (ii) an Evaluation-based Shallow Reflection Module (ESRM), which analyzes negotiation outcomes and provides feedback to guide subsequent negotiations, thereby accelerating consensus formation; and (iii) an LLM-based Deep Reflection Module (LDRM), which performs long-term reflection over negotiation histories to mitigate cognitive fixation and prevent the system from converging to suboptimal solutions. Experimental results in the CARLA simulation environment demonstrate that CoLMIN significantly outperforms existing methods in challenging interactive driving scenarios.