@inproceedings{mesgar-etal-2023-devil,
title = "The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification",
author = "Mesgar, Mohsen and
Tran, Thy Thy and
Glava{\v{s}}, Goran and
Gurevych, Iryna",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eacl-main.135",
doi = "10.18653/v1/2023.eacl-main.135",
pages = "1846--1857",
abstract = "In task-oriented dialog (ToD) new intents emerge on regular basis, with a handful of available utterances at best. This renders effective Few-Shot Intent Classification (FSIC) a central challenge for modular ToD systems. Recent FSIC methods appear to be similar: they use pretrained language models (PLMs) to encode utterances and predominantly resort to nearest-neighbor-based inference. However, they also differ in major components: they start from different PLMs, use different encoding architectures and utterance similarity functions, and adopt different training regimes. Coupling of these vital components together with the lack of informative ablations prevents the identification of factors that drive the (reported) FSIC performance. We propose a unified framework to evaluate these components along the following key dimensions:(1) Encoding architectures: Cross-Encoder vs Bi-Encoders;(2) Similarity function: Parameterized (i.e., trainable) vs non-parameterized; (3) Training regimes: Episodic meta-learning vs conventional (i.e., non-episodic) training. Our experimental results on seven FSIC benchmarks reveal three new important findings. First, the unexplored combination of cross-encoder architecture and episodic meta-learning consistently yields the best FSIC performance. Second, episodic training substantially outperforms its non-episodic counterpart. Finally, we show that splitting episodes into support and query sets has a limited and inconsistent effect on performance. Our findings show the importance of ablations and fair comparisons in FSIC. We publicly release our code and data.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mesgar-etal-2023-devil">
<titleInfo>
<title>The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mohsen</namePart>
<namePart type="family">Mesgar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thy</namePart>
<namePart type="given">Thy</namePart>
<namePart type="family">Tran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Goran</namePart>
<namePart type="family">Glavaš</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iryna</namePart>
<namePart type="family">Gurevych</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In task-oriented dialog (ToD) new intents emerge on regular basis, with a handful of available utterances at best. This renders effective Few-Shot Intent Classification (FSIC) a central challenge for modular ToD systems. Recent FSIC methods appear to be similar: they use pretrained language models (PLMs) to encode utterances and predominantly resort to nearest-neighbor-based inference. However, they also differ in major components: they start from different PLMs, use different encoding architectures and utterance similarity functions, and adopt different training regimes. Coupling of these vital components together with the lack of informative ablations prevents the identification of factors that drive the (reported) FSIC performance. We propose a unified framework to evaluate these components along the following key dimensions:(1) Encoding architectures: Cross-Encoder vs Bi-Encoders;(2) Similarity function: Parameterized (i.e., trainable) vs non-parameterized; (3) Training regimes: Episodic meta-learning vs conventional (i.e., non-episodic) training. Our experimental results on seven FSIC benchmarks reveal three new important findings. First, the unexplored combination of cross-encoder architecture and episodic meta-learning consistently yields the best FSIC performance. Second, episodic training substantially outperforms its non-episodic counterpart. Finally, we show that splitting episodes into support and query sets has a limited and inconsistent effect on performance. Our findings show the importance of ablations and fair comparisons in FSIC. We publicly release our code and data.</abstract>
<identifier type="citekey">mesgar-etal-2023-devil</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.135</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.135</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>1846</start>
<end>1857</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification
%A Mesgar, Mohsen
%A Tran, Thy Thy
%A Glavaš, Goran
%A Gurevych, Iryna
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F mesgar-etal-2023-devil
%X In task-oriented dialog (ToD) new intents emerge on regular basis, with a handful of available utterances at best. This renders effective Few-Shot Intent Classification (FSIC) a central challenge for modular ToD systems. Recent FSIC methods appear to be similar: they use pretrained language models (PLMs) to encode utterances and predominantly resort to nearest-neighbor-based inference. However, they also differ in major components: they start from different PLMs, use different encoding architectures and utterance similarity functions, and adopt different training regimes. Coupling of these vital components together with the lack of informative ablations prevents the identification of factors that drive the (reported) FSIC performance. We propose a unified framework to evaluate these components along the following key dimensions:(1) Encoding architectures: Cross-Encoder vs Bi-Encoders;(2) Similarity function: Parameterized (i.e., trainable) vs non-parameterized; (3) Training regimes: Episodic meta-learning vs conventional (i.e., non-episodic) training. Our experimental results on seven FSIC benchmarks reveal three new important findings. First, the unexplored combination of cross-encoder architecture and episodic meta-learning consistently yields the best FSIC performance. Second, episodic training substantially outperforms its non-episodic counterpart. Finally, we show that splitting episodes into support and query sets has a limited and inconsistent effect on performance. Our findings show the importance of ablations and fair comparisons in FSIC. We publicly release our code and data.
%R 10.18653/v1/2023.eacl-main.135
%U https://aclanthology.org/2023.eacl-main.135
%U https://doi.org/10.18653/v1/2023.eacl-main.135
%P 1846-1857
Markdown (Informal)
[The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification](https://aclanthology.org/2023.eacl-main.135) (Mesgar et al., EACL 2023)
ACL