Policy Improvement using Language Feedback Models
This work addresses the challenge of enhancing imitation learning efficiency and interpretability in language grounding tasks, though it appears incremental as it builds on existing methods like behavioral cloning and LLM feedback.
The paper tackles the problem of improving imitation learning for instruction following by introducing Language Feedback Models (LFMs) that identify desirable behavior from language feedback, resulting in increased task-completion rates over baselines in multiple environments and generalization with 3.5-12.0% improvements after adaptation.
We introduce Language Feedback Models (LFMs) that identify desirable behaviour - actions that help achieve tasks specified in the instruction - for imitation learning in instruction following. To train LFMs, we obtain feedback from Large Language Models (LLMs) on visual trajectories verbalized to language descriptions. First, by using LFMs to identify desirable behaviour to imitate, we improve in task-completion rate over strong behavioural cloning baselines on three distinct language grounding environments (Touchdown, ScienceWorld, and ALFWorld). Second, LFMs outperform using LLMs as experts to directly predict actions, when controlling for the number of LLM output tokens. Third, LFMs generalize to unseen environments, improving task-completion rate by 3.5-12.0% through one round of adaptation. Finally, LFM can be modified to provide human-interpretable feedback without performance loss, allowing human verification of desirable behaviour for imitation learning.