The Yokai Learning Environment: Tracking Beliefs Over Space and Time
This addresses the gap in ToM benchmarks for collaborative AI by providing a dynamic environment to study belief tracking and common ground, though it is incremental in extending existing ToM assessment methods.
The authors tackled the problem of assessing Theory of Mind (ToM) in collaborative AI by introducing the Yokai Learning Environment (YLE), a multi-agent RL environment based on a cooperative card game, and found that current RL agents struggle to solve it even with perfect memory and fail to generalize effectively to unseen partners or longer games.
Developing collaborative AI hinges on Theory of Mind (ToM) - the ability to reason about the beliefs of others to build and maintain common ground. Existing ToM benchmarks, however, are restricted to passive observer settings or lack an assessment of how agents establish and maintain common ground over time. To address these gaps, we introduce the Yokai Learning Environment (YLE) - a multi-agent reinforcement learning (RL) environment based on the cooperative card game Yokai. In the YLE, agents take turns peeking at hidden cards and moving them to form clusters based on colour. Success requires tracking evolving beliefs, remembering past observations, using hints as grounded communication, and maintaining common ground with teammates. Our evaluation yields two key findings: First, current RL agents struggle to solve the YLE, even when given access to perfect memory. Second, while belief modelling improves performance, agents are still unable to effectively generalise to unseen partners or form accurate beliefs over longer games, exposing a reliance on brittle conventions rather than robust belief tracking. We use the YLE to investigate research questions in belief modelling, memory, partner generalisation, and scaling to higher-order ToM.