From Disfluency Detection to Intent Detection and Slot Filling
This addresses the problem of handling disfluencies in spoken language understanding for low-resource languages like Vietnamese, but it is incremental as it extends existing methods to a new dataset.
The study investigated how disfluency detection affects intent detection and slot filling in Vietnamese, a low-resource language, by creating a disfluency-annotated dataset and finding that disfluencies negatively impact performance, with XLM-R outperforming PhoBERT in disfluent contexts.
We present the first empirical study investigating the influence of disfluency detection on downstream tasks of intent detection and slot filling. We perform this study for Vietnamese -- a low-resource language that has no previous study as well as no public dataset available for disfluency detection. First, we extend the fluent Vietnamese intent detection and slot filling dataset PhoATIS by manually adding contextual disfluencies and annotating them. Then, we conduct experiments using strong baselines for disfluency detection and joint intent detection and slot filling, which are based on pre-trained language models. We find that: (i) disfluencies produce negative effects on the performances of the downstream intent detection and slot filling tasks, and (ii) in the disfluency context, the pre-trained multilingual language model XLM-R helps produce better intent detection and slot filling performances than the pre-trained monolingual language model PhoBERT, and this is opposite to what generally found in the fluency context.