@inproceedings{van-de-kar-etal-2022-dont,
title = "Don`t Prompt, Search! Mining-based Zero-Shot Learning with Language Models",
author = "van de Kar, Mozes and
Xia, Mengzhou and
Chen, Danqi and
Artetxe, Mikel",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.509/",
doi = "10.18653/v1/2022.emnlp-main.509",
pages = "7508--7520",
abstract = "Masked language models like BERT can perform text classification in a zero-shot fashion by reformulating downstream tasks as text infilling. However, this approach is highly sensitive to the template used to prompt the model, yet practitioners are blind when designing them in strict zero-shot settings. In this paper, we propose an alternative mining-based approach for zero-shot learning. Instead of prompting language models, we use regular expressions to mine labeled examples from unlabeled corpora, which can optionally be filtered through prompting, and used to finetune a pretrained model. Our method is more flexible and interpretable than prompting, and outperforms it on a wide range of tasks when using comparable templates. Our results suggest that the success of prompting can partly be explained by the model being exposed to similar examples during pretraining, which can be directly retrieved through regular expressions."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="van-de-kar-etal-2022-dont">
<titleInfo>
<title>Don‘t Prompt, Search! Mining-based Zero-Shot Learning with Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mozes</namePart>
<namePart type="family">van de Kar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengzhou</namePart>
<namePart type="family">Xia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Danqi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mikel</namePart>
<namePart type="family">Artetxe</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>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</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>Masked language models like BERT can perform text classification in a zero-shot fashion by reformulating downstream tasks as text infilling. However, this approach is highly sensitive to the template used to prompt the model, yet practitioners are blind when designing them in strict zero-shot settings. In this paper, we propose an alternative mining-based approach for zero-shot learning. Instead of prompting language models, we use regular expressions to mine labeled examples from unlabeled corpora, which can optionally be filtered through prompting, and used to finetune a pretrained model. Our method is more flexible and interpretable than prompting, and outperforms it on a wide range of tasks when using comparable templates. Our results suggest that the success of prompting can partly be explained by the model being exposed to similar examples during pretraining, which can be directly retrieved through regular expressions.</abstract>
<identifier type="citekey">van-de-kar-etal-2022-dont</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.509</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.509/</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>7508</start>
<end>7520</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Don‘t Prompt, Search! Mining-based Zero-Shot Learning with Language Models
%A van de Kar, Mozes
%A Xia, Mengzhou
%A Chen, Danqi
%A Artetxe, Mikel
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F van-de-kar-etal-2022-dont
%X Masked language models like BERT can perform text classification in a zero-shot fashion by reformulating downstream tasks as text infilling. However, this approach is highly sensitive to the template used to prompt the model, yet practitioners are blind when designing them in strict zero-shot settings. In this paper, we propose an alternative mining-based approach for zero-shot learning. Instead of prompting language models, we use regular expressions to mine labeled examples from unlabeled corpora, which can optionally be filtered through prompting, and used to finetune a pretrained model. Our method is more flexible and interpretable than prompting, and outperforms it on a wide range of tasks when using comparable templates. Our results suggest that the success of prompting can partly be explained by the model being exposed to similar examples during pretraining, which can be directly retrieved through regular expressions.
%R 10.18653/v1/2022.emnlp-main.509
%U https://aclanthology.org/2022.emnlp-main.509/
%U https://doi.org/10.18653/v1/2022.emnlp-main.509
%P 7508-7520
Markdown (Informal)
[Don’t Prompt, Search! Mining-based Zero-Shot Learning with Language Models](https://aclanthology.org/2022.emnlp-main.509/) (van de Kar et al., EMNLP 2022)
ACL