Baba Is AI: Break the Rules to Beat the Benchmark
This addresses the challenge of assessing creative rule-breaking in AI for researchers and developers, though it is incremental as it introduces a new benchmark rather than a solution.
The researchers tackled the problem of evaluating AI agents' ability to manipulate and combine rules in problem-solving by creating a new benchmark based on the game Baba Is You, and found that state-of-the-art multi-modal large language models like GPT-4o and Gemini-1.5 fail dramatically in this task.
Humans solve problems by following existing rules and procedures, and also by leaps of creativity to redefine those rules and objectives. To probe these abilities, we developed a new benchmark based on the game Baba Is You where an agent manipulates both objects in the environment and rules, represented by movable tiles with words written on them, to reach a specified goal and win the game. We test three state-of-the-art multi-modal large language models (OpenAI GPT-4o, Google Gemini-1.5-Pro and Gemini-1.5-Flash) and find that they fail dramatically when generalization requires that the rules of the game must be manipulated and combined.