Teach Me What to Say and I Will Learn What to Pick: Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models

Ehsan Lotfi, Maxime De Bruyn, Jeska Buhmann, Walter Daelemans


Abstract
Knowledge Grounded Conversation Models are usually based on a selection/retrieval module and a generation module, trained separately or simultaneously, with or without having access to a ‘gold’ knowledge option. With the introduction of large pre-trained generative models, the selection and generation part have become more and more entangled, shifting the focus towards enhancing knowledge incorporation (from multiple sources) instead of trying to pick the best knowledge option. These approaches however depend on knowledge labels and/or a separate dense retriever for their best performance. In this work we study the unsupervised selection abilities of pre-trained generative models (e.g. BART) and show that by adding a score-and-aggregate module between encoder and decoder, they are capable of learning to pick the proper knowledge through minimising the language modelling loss (i.e. without having access to knowledge labels). Trained as such, our model - K-Mine - shows competitive selection and generation performance against models that benefit from knowledge labels and/or separate dense retriever.
Anthology ID:
2021.nlp4convai-1.24
Volume:
Proceedings of the 3rd Workshop on Natural Language Processing for Conversational AI
Month:
November
Year:
2021
Address:
Online
Venues:
EMNLP | NLP4ConvAI
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
254–262
Language:
URL:
https://aclanthology.org/2021.nlp4convai-1.24
DOI:
10.18653/v1/2021.nlp4convai-1.24
Bibkey:
Cite (ACL):
Ehsan Lotfi, Maxime De Bruyn, Jeska Buhmann, and Walter Daelemans. 2021. Teach Me What to Say and I Will Learn What to Pick: Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models. In Proceedings of the 3rd Workshop on Natural Language Processing for Conversational AI, pages 254–262, Online. Association for Computational Linguistics.
Cite (Informal):
Teach Me What to Say and I Will Learn What to Pick: Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models (Lotfi et al., NLP4ConvAI 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.nlp4convai-1.24.pdf
Data
Holl-EWizard of Wikipedia