Search papers, labs, and topics across Lattice.
This paper introduces UniETP, a unified interface that integrates four popular simulators for Embodied Task Planning, addressing the fragmentation in existing datasets and simulators that complicates model evaluation. By standardizing observation and action spaces while enhancing task diversity and complexity, UniETP facilitates comprehensive evaluation of embodied planning models across varied task domains. Extensive experiments reveal performance bottlenecks in current models, highlighting the need for improved generalization in embodied task execution.
UniETP reveals that standardizing diverse simulators can significantly enhance the evaluation and scalability of embodied task planning models.
This paper focuses on the problem of Embodied Task Planning, where an agent is required to execute a sequence of atomic actions within an interactive environment to complete a user-specified task. Though a variety of simulators and datasets have previously been built for this task, these efforts are largely isolated, with each using its own observation format, action type, and task domain. This fragmentation complicates comprehensive model evaluation and hinders the scalability of training data. As an effort towards generalizable embodied planning, we propose UniETP, a unified interface integrating four commonly-used simulators (AI2-THOR, VirtualHome, Habitat, BEHAVIOR). UniETP is characterized by both standardization and diversity. On one hand, it formalizes all the simulators into a consistent observation and action space, and builds an evaluation system to support complicated task goal. On the other hand, it enhances task diversity and complexity across dimensions like task logic, instance grounding, and instruction understanding, constructing a new dataset with varied levels of difficulty in an automatic manner. Extensive experiments on the proposed benchmark are conducted to evaluate the embodied planning capabilities of recent models and analyze the performance bottlenecks. Codes and data will be available at https://github.com/woyut/UniETP .
