Continual Learning for Grounded Instruction Generation by Observing Human Following Behavior
This work addresses the challenge of effective human-AI collaboration through continual learning in instruction generation, though it appears incremental as it builds on existing contextual bandit methods.
The paper tackles the problem of improving a system's ability to generate natural language instructions by observing human execution behavior, using contextual bandit learning to refine instruction generation over time, resulting in dramatic improvements in language generation capabilities.
We study continual learning for natural language instruction generation, by observing human users' instruction execution. We focus on a collaborative scenario, where the system both acts and delegates tasks to human users using natural language. We compare user execution of generated instructions to the original system intent as an indication to the system's success communicating its intent. We show how to use this signal to improve the system's ability to generate instructions via contextual bandit learning. In interaction with real users, our system demonstrates dramatic improvements in its ability to generate language over time.