Factorizing Perception and Policy for Interactive Instruction Following
This work aims to improve AI agents' ability to follow interactive instructions for household tasks, which is an incremental step towards more capable robotic assistants.
This paper addresses the challenge of AI agents performing household tasks based on language instructions. The authors propose MOCA, a Modular Object-Centric Approach, which factorizes the task into interactive perception and action policy streams, achieving significant performance improvements and better generalization on the ALFRED benchmark.
Performing simple household tasks based on language directives is very natural to humans, yet it remains an open challenge for AI agents. The 'interactive instruction following' task attempts to make progress towards building agents that jointly navigate, interact, and reason in the environment at every step. To address the multifaceted problem, we propose a model that factorizes the task into interactive perception and action policy streams with enhanced components and name it as MOCA, a Modular Object-Centric Approach. We empirically validate that MOCA outperforms prior arts by significant margins on the ALFRED benchmark with improved generalization.