研究世界模型的效率与可解释性权衡，为智能体安全测试提供设计指南。
AI in a vat: Fundamental limits of efficient world modelling for agent sandboxing and interpretability
- 基于计算力学原理，简化世界模型以适应不同评估需求。
- 揭示高效建模与可解释性不可兼得的根本矛盾。
- 适用于需要安全测试和可解释性的智能体系统设计者。
近期工作提出使用世界模型构建受控虚拟环境，以在部署前测试人工智能智能体的可靠性与安全性。然而，高精度世界模型通常计算开销大，严重限制了评估的广度与深度。受经典‘缸中之脑’思想实验启发，本文研究如何简化世界模型，且不依赖被评估智能体的具体特性。基于计算力学原则，我们的方法揭示了世界模型构建中效率与可解释性之间的根本权衡，证明不存在能同时优化所有理想特性的单一模型。基于此权衡，我们提出了三种构建策略：最小化内存占用、界定可学习范围，以及追踪不良结果的成因。该研究确立了世界建模的基本限制，提供了可操作的设计指导，有助于智能体评估的核心决策。
原文摘要 · Abstract (English)
Recent work proposes using world models to generate controlled virtual environments in which AI agents can be tested before deployment to ensure their reliability and safety. However, accurate world models often have high computational demands that can severely restrict the scope and depth of such assessments. Inspired by the classic `brain in a vat' thought experiment, here we investigate ways of simplifying world models that remain agnostic to the AI agent under evaluation. By following principles from computational mechanics, our approach reveals a fundamental trade-off in world model construction between efficiency and interpretability, demonstrating that no single world model can optimise all desirable characteristics. Building on this trade-off, we identify procedures to build world models that either minimise memory requirements, delineate the boundaries of what is learnable, or allow tracking causes of undesirable outcomes. In doing so, this work establishes fundamental limits in world modelling, leading to actionable guidelines that inform core design choices related to effective agent evaluation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。