TY - GEN
T1 - Selective In-Context Data Augmentation for Intent Detection using Pointwise V-Information
AU - Lin, Yen Ting
AU - Papangelis, Alexandros
AU - Kim, Seokhwan
AU - Lee, Sungjin
AU - Hazarika, Devamanyu
AU - Namazifar, Mahdi
AU - Jin, Di
AU - Liu, Yang
AU - Hakkani-Tur, Dilek
N1 - Publisher Copyright:
© 2023 Association for Computational Linguistics.
PY - 2023
Y1 - 2023
N2 - This work focuses on in-context data augmentation for intent detection. Having found that augmentation via in-context prompting of large pretrained language models (PLMs) alone does not improve performance, we introduce a novel approach based on PLMs and pointwise V-information (PVI), a metric that can measure the usefulness of a datapoint for training a model. Our method first fine-tunes a PLM on a small seed of training data and then synthesizes new datapoints - utterances that correspond to given intents. It then employs intent-aware filtering, based on PVI, to remove datapoints that are not helpful to the downstream intent classifier. Our method is thus able to leverage the expressive power of large language models to produce diverse training data. Empirical results demonstrate that our method can produce synthetic training data that achieve state-of-the-art performance on three challenging intent detection datasets under few-shot settings (1.28% absolute improvement in 5-shot and 1.18% absolute in 10-shot, on average) and perform on par with the state-of-the-art in full-shot settings (within 0.01% absolute, on average).
AB - This work focuses on in-context data augmentation for intent detection. Having found that augmentation via in-context prompting of large pretrained language models (PLMs) alone does not improve performance, we introduce a novel approach based on PLMs and pointwise V-information (PVI), a metric that can measure the usefulness of a datapoint for training a model. Our method first fine-tunes a PLM on a small seed of training data and then synthesizes new datapoints - utterances that correspond to given intents. It then employs intent-aware filtering, based on PVI, to remove datapoints that are not helpful to the downstream intent classifier. Our method is thus able to leverage the expressive power of large language models to produce diverse training data. Empirical results demonstrate that our method can produce synthetic training data that achieve state-of-the-art performance on three challenging intent detection datasets under few-shot settings (1.28% absolute improvement in 5-shot and 1.18% absolute in 10-shot, on average) and perform on par with the state-of-the-art in full-shot settings (within 0.01% absolute, on average).
UR - http://www.scopus.com/inward/record.url?scp=85159857144&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85159857144&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85159857144
T3 - EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference
SP - 1455
EP - 1468
BT - EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference
PB - Association for Computational Linguistics (ACL)
T2 - 17th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2023
Y2 - 2 May 2023 through 6 May 2023
ER -