Generating Repetitions with Appropriate Repeated Words
This addresses a specific linguistic challenge in dialogue systems for improving communication trust, but it is incremental as it builds on existing pre-trained models.
The paper tackled the problem of generating repetitions in dialogue, which are responses that repeat words from previous utterances to build trust, by proposing Weighted Label Smoothing and a repetition scoring method applied to T5. The results showed that these methods outperformed baselines in both automatic and human evaluations.
A repetition is a response that repeats words in the previous speaker's utterance in a dialogue. Repetitions are essential in communication to build trust with others, as investigated in linguistic studies. In this work, we focus on repetition generation. To the best of our knowledge, this is the first neural approach to address repetition generation. We propose Weighted Label Smoothing, a smoothing method for explicitly learning which words to repeat during fine-tuning, and a repetition scoring method that can output more appropriate repetitions during decoding. We conducted automatic and human evaluations involving applying these methods to the pre-trained language model T5 for generating repetitions. The experimental results indicate that our methods outperformed baselines in both evaluations.