Evaluating Generalization Capabilities of LLM-Based Agents in Mixed-Motive Scenarios Using Concordia

TL;DR

This paper introduces a new evaluation method using Concordia to assess how well large language model agents can generalize cooperative behavior across diverse social scenarios, highlighting current limitations.

Contribution

It presents a novel zero-shot evaluation framework for LLM agents in mixed-motive environments, emphasizing generalization in social cooperation tasks.

Findings

Agents show limited generalization in diverse scenarios

Significant gaps in cooperation skills, especially in persuasion

Current agents struggle with norm enforcement

Abstract

Large Language Model (LLM) agents have demonstrated impressive capabilities for social interaction and are increasingly being deployed in situations where they might engage with both human and artificial agents. These interactions represent a critical frontier for LLM-based agents, yet existing evaluation methods fail to measure how well these capabilities generalize to novel social situations. In this paper, we introduce a method for evaluating the ability of LLM-based agents to cooperate in zero-shot, mixed-motive environments using Concordia, a natural language multi-agent simulation environment. Our method measures general cooperative intelligence by testing an agent's ability to identify and exploit opportunities for mutual gain across diverse partners and contexts. We present empirical results from the NeurIPS 2024 Concordia Contest, where agents were evaluated on their ability to achieve mutual gains across a suite of diverse scenarios ranging from negotiation to collective action problems. Our findings reveal significant gaps between current agent capabilities and the robust generalization required for reliable cooperation, particularly in scenarios demanding persuasion and norm enforcement.