Taking Action Towards Graceful Interaction: The Effects of Performing Actions on Modelling Policies for Instruction Clarification Requests

Anonymous

Abstract: Clarification requests are a mechanism to help solve communication problems in instruction-following interactions. Despite their importance, even skilful models struggle with producing or interpreting such repair acts. In this work, we show that even well-motivated, Transformer-based models fail to learn a good policy for when to ask Instruction CRs (iCRs), while the task of determining what to ask about can be more successfully predicted. We test three hypotheses concerning the effects of action taking as an auxiliary task for iCR policies, concluding that, while its contribution is limited, some information can be extracted from prediction uncertainty. Considering the implications of these findings, we further discuss the shortcomings of the data-driven paradigm for learning meta-communication acts.

Paper Type: long

Research Area: Dialogue and Interactive Systems

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Publicly available software and/or pre-trained models

Languages Studied: English

Consent To Share Submission Details: On behalf of all authors, we agree to the terms above to share our submission details.