Current data-driven Dialogue State Tracking (DST) models exhibit a poor capacity to adapt themselves to domain changes, resulting in a significant degradation in performance. We propose a methodology, called Generative Dialogue Domain Adaptation, which significantly simplifies the creation of training data when a number of changes (e.g., new slot-values or new instances) occur in a domain Knowledge Base. We start from dialogues for a source domain and apply generative methods based on language models such as BERT, fine-tuned on task-related data and generate slot-values substitutions for a target domain. We have experimented dialogue domain adaptation in a few-shot setting showing promising results, although the task is still very challenging. We provide a deep analysis of the quality of the generated data and of the features that affect this task, and we emphasise that DST models are very sensitive to the distribution of slot-values in the corpus.

Fine-Tuning BERT for Generative Dialogue Domain Adaptation

Tiziano Labruna
Membro del Collaboration Group
;
Bernardo Magnini
Membro del Collaboration Group
2022-01-01

Abstract

Current data-driven Dialogue State Tracking (DST) models exhibit a poor capacity to adapt themselves to domain changes, resulting in a significant degradation in performance. We propose a methodology, called Generative Dialogue Domain Adaptation, which significantly simplifies the creation of training data when a number of changes (e.g., new slot-values or new instances) occur in a domain Knowledge Base. We start from dialogues for a source domain and apply generative methods based on language models such as BERT, fine-tuned on task-related data and generate slot-values substitutions for a target domain. We have experimented dialogue domain adaptation in a few-shot setting showing promising results, although the task is still very challenging. We provide a deep analysis of the quality of the generated data and of the features that affect this task, and we emphasise that DST models are very sensitive to the distribution of slot-values in the corpus.
2022
978-3-031-16269-5
978-3-031-16270-1
File in questo prodotto:
File Dimensione Formato  
_accepted__TSD_2022.pdf

solo utenti autorizzati

Tipologia: Documento in Post-print
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 570.68 kB
Formato Adobe PDF
570.68 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11582/335627
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact