Collab-Overcooked: Benchmarking and Evaluating Large Language Models as Collaborative Agents
This work addresses the need for better evaluation of LLM-based multi-agent systems in interactive environments, though it is incremental as it builds on existing benchmarks.
The authors introduced Collab-Overcooked, a new benchmark for evaluating large language models as collaborative agents in multi-agent systems, based on the Overcooked-AI game, and found that while LLMs excel at goal interpretation, they show significant weaknesses in active collaboration and continuous adaptation.
Large Language Models (LLMs) based agent systems have made great strides in real-world applications beyond traditional NLP tasks. This paper proposes a new LLM-based Multi-Agent System (LLM-MAS) benchmark, Collab-Overcooked, built on the popular Overcooked-AI game with more applicable and challenging tasks in interactive environments. Collab-Overcooked extends existing benchmarks in two novel ways. First, it provides a multi-agent framework supporting diverse tasks and objectives and encourages collaboration through natural language communication. Second, it introduces a spectrum of process-oriented evaluation metrics to assess the fine-grained collaboration capabilities of different LLM agents, a dimension often overlooked in prior work. We conduct extensive experiments with 13 popular LLMs and show that, while the LLMs exhibit a strong ability in goal interpretation, there are significant shortcomings in active collaboration and continuous adaptation, which are critical for efficiently fulfilling complex tasks. Notably, we highlight the strengths and weaknesses of LLM-MAS and provide insights for improving and evaluating LLM-MAS on a unified and open-source benchmark. The environments, 30 open-ended tasks, and the evaluation package are publicly available at https://github.com/YusaeMeow/Collab-Overcooked.