AnyGoal: On World Model Generalization Beyond Expert Rollouts· AnyGoal：超越专家轨迹的世界模型泛化
The goal of a world model is to capture the dynamics of an environment. This has the great appeal that it allows agents to plan and simulate a course of actions with the help of a world model. For these uses to be reliable, however, the learned dynamics must behave well beyond the specific state-action combinations observed in expert rollouts. Despite this, the relationship between data coverage, dynamics generalization, and downstream planning performance remains underexplored. We study this relationship through the lens of compositional generalization: can a world model recombine familiar state and action factors into valid dynamics? To answer this, we introduce AnyGoal, a benchmark of goal-shifted task variants built from widely used world-model environments, together with an open-sourc
研究世界模型如何泛化到未见状态及动作，提出 AnyGoal 基准。
- 核心方法
- 通过构建AnyGoal这一目标平移任务变体基准，利用广泛使用的世界模型环境，研究世界模型的组合泛化能力，即能否重新组合已知的状态和动作因子形成有效的动态。
- 适合谁读
- 研究者
- 要解决的问题
- 现有世界模型在未见状态动作组合上的泛化能力不足，影响了其在下游任务中的表现。
- 关键实验
- 使用了广泛认可的世界模型环境构建了多种任务变体进行实验，具体数据未详细提供。
- 主要贡献
- 提出了AnyGoal新基准，促进了对世界模型泛化能力尤其是组合泛化能力的研究。
- 意义与局限
- 加深了对世界模型泛化机制的理解，为提高模型在未知环境中的可靠性和适应性提供了新思路；然而，该研究专注于特定类型的世界模型环境，可能无法全面反映所有场景。