@inproceedings{parikh-etal-2023-exploring,
title = "Exploring Zero and Few-shot Techniques for Intent Classification",
author = "Parikh, Soham and
Tiwari, Mitul and
Tumbade, Prashil and
Vohra, Quaizar",
editor = "Sitaram, Sunayana and
Beigman Klebanov, Beata and
Williams, Jason D",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-industry.71",
doi = "10.18653/v1/2023.acl-industry.71",
pages = "744--751",
abstract = "Conversational NLU providers often need to scale to thousands of intent-classification models where new customers often face the cold-start problem. Scaling to so many customers puts a constraint on storage space as well. In this paper, we explore four different zero and few-shot intent classification approaches with this low-resource constraint: 1) domain adaptation, 2) data augmentation, 3) zero-shot intent classification using descriptions large language models (LLMs), and 4) parameter-efficient fine-tuning of instruction-finetuned language models. Our results show that all these approaches are effective to different degrees in low-resource settings. Parameter-efficient fine-tuning using T-few recipe on Flan-T5 yields the best performance even with just one sample per intent. We also show that the zero-shot method of prompting LLMs using intent descriptions is also very competitive.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="parikh-etal-2023-exploring">
<titleInfo>
<title>Exploring Zero and Few-shot Techniques for Intent Classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Soham</namePart>
<namePart type="family">Parikh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mitul</namePart>
<namePart type="family">Tiwari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prashil</namePart>
<namePart type="family">Tumbade</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Quaizar</namePart>
<namePart type="family">Vohra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sunayana</namePart>
<namePart type="family">Sitaram</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Beata</namePart>
<namePart type="family">Beigman Klebanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jason</namePart>
<namePart type="given">D</namePart>
<namePart type="family">Williams</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Conversational NLU providers often need to scale to thousands of intent-classification models where new customers often face the cold-start problem. Scaling to so many customers puts a constraint on storage space as well. In this paper, we explore four different zero and few-shot intent classification approaches with this low-resource constraint: 1) domain adaptation, 2) data augmentation, 3) zero-shot intent classification using descriptions large language models (LLMs), and 4) parameter-efficient fine-tuning of instruction-finetuned language models. Our results show that all these approaches are effective to different degrees in low-resource settings. Parameter-efficient fine-tuning using T-few recipe on Flan-T5 yields the best performance even with just one sample per intent. We also show that the zero-shot method of prompting LLMs using intent descriptions is also very competitive.</abstract>
<identifier type="citekey">parikh-etal-2023-exploring</identifier>
<identifier type="doi">10.18653/v1/2023.acl-industry.71</identifier>
<location>
<url>https://aclanthology.org/2023.acl-industry.71</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>744</start>
<end>751</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Exploring Zero and Few-shot Techniques for Intent Classification
%A Parikh, Soham
%A Tiwari, Mitul
%A Tumbade, Prashil
%A Vohra, Quaizar
%Y Sitaram, Sunayana
%Y Beigman Klebanov, Beata
%Y Williams, Jason D.
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F parikh-etal-2023-exploring
%X Conversational NLU providers often need to scale to thousands of intent-classification models where new customers often face the cold-start problem. Scaling to so many customers puts a constraint on storage space as well. In this paper, we explore four different zero and few-shot intent classification approaches with this low-resource constraint: 1) domain adaptation, 2) data augmentation, 3) zero-shot intent classification using descriptions large language models (LLMs), and 4) parameter-efficient fine-tuning of instruction-finetuned language models. Our results show that all these approaches are effective to different degrees in low-resource settings. Parameter-efficient fine-tuning using T-few recipe on Flan-T5 yields the best performance even with just one sample per intent. We also show that the zero-shot method of prompting LLMs using intent descriptions is also very competitive.
%R 10.18653/v1/2023.acl-industry.71
%U https://aclanthology.org/2023.acl-industry.71
%U https://doi.org/10.18653/v1/2023.acl-industry.71
%P 744-751
Markdown (Informal)
[Exploring Zero and Few-shot Techniques for Intent Classification](https://aclanthology.org/2023.acl-industry.71) (Parikh et al., ACL 2023)
ACL
- Soham Parikh, Mitul Tiwari, Prashil Tumbade, and Quaizar Vohra. 2023. Exploring Zero and Few-shot Techniques for Intent Classification. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track), pages 744–751, Toronto, Canada. Association for Computational Linguistics.
