Overcooked-AI

Overcooked-AI is a benchmark that contains scenarios from the cooperative cooking game "Overcooked," used to evaluate human-AI collaboration and the effectiveness of policies in multi-agent environments.

Papers using Overcooked-AI (4)

Collab-overcooked: Benchmarking And Evaluating Large Language Models As Collaborative Agents2025Large Language Model Guided Incentive Aware Reward Design for Cooperative Multi-Agent Reinforcement Learning2026Generalizable Agent Modeling for Agent Collaboration-Competition Adaptation with Multi-Retrieval and Dynamic Generation2025Towards Real-time Adaptation Of Embodied Agent In Human-robot Collaboration2024