@inproceedings{jiang-etal-2022-improved,
title = "Improved Universal Sentence Embeddings with Prompt-based Contrastive Learning and Energy-based Learning",
author = "Jiang, Yuxin and
Zhang, Linhan and
Wang, Wei",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.220",
doi = "10.18653/v1/2022.findings-emnlp.220",
pages = "3021--3035",
abstract = "Contrastive learning has been demonstrated to be effective in enhancing pre-trained language models (PLMs) to derive superior universal sentence embeddings. However, existing contrastive methods still have two limitations. Firstly, previous works may acquire poor performance under domain shift settings, thus hindering the application of sentence representations in practice. We attribute this low performance to the over-parameterization of PLMs with millions of parameters. To alleviate it, we propose PromCSE (Prompt-based Contrastive Learning for Sentence Embeddings), which only trains small-scale Soft Prompt (i.e., a set of trainable vectors) while keeping PLMs fixed. Secondly, the commonly used NT-Xent loss function of contrastive learning does not fully exploit hard negatives in supervised learning settings. To this end, we propose to integrate an Energy-based Hinge loss to enhance the pairwise discriminative power, inspired by the connection between the NT-Xent loss and the Energy-based Learning paradigm. Empirical results on seven standard semantic textual similarity (STS) tasks and a domain-shifted STS task both show the effectiveness of our method compared with the current state-of-the-art sentence embedding models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jiang-etal-2022-improved">
<titleInfo>
<title>Improved Universal Sentence Embeddings with Prompt-based Contrastive Learning and Energy-based Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuxin</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linhan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Contrastive learning has been demonstrated to be effective in enhancing pre-trained language models (PLMs) to derive superior universal sentence embeddings. However, existing contrastive methods still have two limitations. Firstly, previous works may acquire poor performance under domain shift settings, thus hindering the application of sentence representations in practice. We attribute this low performance to the over-parameterization of PLMs with millions of parameters. To alleviate it, we propose PromCSE (Prompt-based Contrastive Learning for Sentence Embeddings), which only trains small-scale Soft Prompt (i.e., a set of trainable vectors) while keeping PLMs fixed. Secondly, the commonly used NT-Xent loss function of contrastive learning does not fully exploit hard negatives in supervised learning settings. To this end, we propose to integrate an Energy-based Hinge loss to enhance the pairwise discriminative power, inspired by the connection between the NT-Xent loss and the Energy-based Learning paradigm. Empirical results on seven standard semantic textual similarity (STS) tasks and a domain-shifted STS task both show the effectiveness of our method compared with the current state-of-the-art sentence embedding models.</abstract>
<identifier type="citekey">jiang-etal-2022-improved</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.220</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.220</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3021</start>
<end>3035</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improved Universal Sentence Embeddings with Prompt-based Contrastive Learning and Energy-based Learning
%A Jiang, Yuxin
%A Zhang, Linhan
%A Wang, Wei
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F jiang-etal-2022-improved
%X Contrastive learning has been demonstrated to be effective in enhancing pre-trained language models (PLMs) to derive superior universal sentence embeddings. However, existing contrastive methods still have two limitations. Firstly, previous works may acquire poor performance under domain shift settings, thus hindering the application of sentence representations in practice. We attribute this low performance to the over-parameterization of PLMs with millions of parameters. To alleviate it, we propose PromCSE (Prompt-based Contrastive Learning for Sentence Embeddings), which only trains small-scale Soft Prompt (i.e., a set of trainable vectors) while keeping PLMs fixed. Secondly, the commonly used NT-Xent loss function of contrastive learning does not fully exploit hard negatives in supervised learning settings. To this end, we propose to integrate an Energy-based Hinge loss to enhance the pairwise discriminative power, inspired by the connection between the NT-Xent loss and the Energy-based Learning paradigm. Empirical results on seven standard semantic textual similarity (STS) tasks and a domain-shifted STS task both show the effectiveness of our method compared with the current state-of-the-art sentence embedding models.
%R 10.18653/v1/2022.findings-emnlp.220
%U https://aclanthology.org/2022.findings-emnlp.220
%U https://doi.org/10.18653/v1/2022.findings-emnlp.220
%P 3021-3035
Markdown (Informal)
[Improved Universal Sentence Embeddings with Prompt-based Contrastive Learning and Energy-based Learning](https://aclanthology.org/2022.findings-emnlp.220) (Jiang et al., Findings 2022)
ACL