Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest

TL;DR

This paper introduces C2C, a multi-agent environment for studying negotiation and cooperation among AI agents and humans, revealing behavioral differences and improving agent performance through targeted prompting.

Contribution

It presents a new environment for testing LM-based agents in mixed-motive negotiations, analyzes human versus AI behaviors, and enhances agent success rates with behavioral prompts.

Findings

Humans prefer simpler deals and are less reliable partners.

Humans accept deals without counteroffers 56.3% of the time, compared to 67.6% for AI.

Targeted prompting increased AI win rates from 22.2% to 32.7%.

Abstract

Language Model (LM)-based agents remain largely untested in mixed-motive settings where agents must leverage short-term cooperation for long-term competitive goals (e.g., multi-party politics). We introduce Cooperate to Compete (C2C), a multi-agent environment where players can engage in private negotiations while competing to be the first to achieve their secret objective. Players have asymmetric objectives and negotiations are non-binding, allowing alliances to form and break as players' short-term interests align and diverge. We run AI only games and conduct a user study pitting human players against AI opponents. We identify significant differences between human and AI negotiation behaviors, finding that humans favor lower-complexity deals and are significantly less reliable partners compared to LM-based agents. We also find that humans are more aggressive negotiators, accepting deals without a counteroffer only 56.3% of the time compared to 67.6% for LM-based agents. Through targeted prompting inspired by these findings, we modify agents' negotiation behavior and improve win rates from 22.2% to 32.7%. We run over 1,100 games with over 16,000 private conversations totaling 15.2 million tokens and over 150,000 player actions. Our results establish C2C as a testbed for studying and building LM-based agents that can navigate the sophisticated coordination required for real-world deployments. The game, code, and dataset may be found at https://negotiationgame.io/c2c.