How About Kind of Generating Hedges using End-to-End Neural Models?
This work addresses the challenge of softening statements to avoid face threat in noisy tutoring contexts, but it is incremental as it builds on existing models and methods.
The authors tackled the problem of generating hedges in noisy peer-tutoring conversations by fine-tuning state-of-the-art language models and using reranking with a hedge classifier, showing that generation is feasible in this environment.
Hedging is a strategy for softening the impact of a statement in conversation. In reducing the strength of an expression, it may help to avoid embarrassment (more technically, ``face threat'') to one's listener. For this reason, it is often found in contexts of instruction, such as tutoring. In this work, we develop a model of hedge generation based on i) fine-tuning state-of-the-art language models trained on human-human tutoring data, followed by ii) reranking to select the candidate that best matches the expected hedging strategy within a candidate pool using a hedge classifier. We apply this method to a natural peer-tutoring corpus containing a significant number of disfluencies, repetitions, and repairs. The results show that generation in this noisy environment is feasible with reranking. By conducting an error analysis for both approaches, we reveal the challenges faced by systems attempting to accomplish both social and task-oriented goals in conversation.