<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="schick-schutze-2022-true">
    <titleInfo>
        <title>True Few-Shot Learning with Prompts—A Real-World Perspective</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Timo</namePart>
        <namePart type="family">Schick</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hinrich</namePart>
        <namePart type="family">Schütze</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Prompt-based approaches excel at few-shot learning. However, Perez et al. (2021) recently cast doubt on their performance as they had difficulty getting good results in a “true” few-shot setting in which prompts and hyperparameters cannot be tuned on a dev set. In view of this, we conduct an extensive study of Pet, a method that combines textual instructions with example-based finetuning. We show that, if correctly configured, Pet performs strongly in true few-shot settings without a dev set. Crucial for this strong performance is a number of design choices, including Pet‘s ability to intelligently handle multiple prompts. We put our findings to a real-world test by running Pet on RAFT, a benchmark of tasks taken from realistic NLP applications for which no labeled dev or test sets are available. Pet achieves a new state of the art on RAFT and performs close to non-expert humans for 7 out of 11 tasks. These results demonstrate that prompt-based learners can successfully be applied in true few-shot settings and underpin our belief that learning from instructions will play an important role on the path towards human-like few-shot learning capabilities.</abstract>
    <identifier type="citekey">schick-schutze-2022-true</identifier>
    <identifier type="doi">10.1162/tacl_a_00485</identifier>
    <location>
        <url>https://aclanthology.org/2022.tacl-1.41/</url>
    </location>
    <part>
        <date>2022</date>
        <detail type="volume"><number>10</number></detail>
        <extent unit="page">
            <start>716</start>
            <end>731</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T True Few-Shot Learning with Prompts—A Real-World Perspective
%A Schick, Timo
%A Schütze, Hinrich
%J Transactions of the Association for Computational Linguistics
%D 2022
%V 10
%I MIT Press
%C Cambridge, MA
%F schick-schutze-2022-true
%X Prompt-based approaches excel at few-shot learning. However, Perez et al. (2021) recently cast doubt on their performance as they had difficulty getting good results in a “true” few-shot setting in which prompts and hyperparameters cannot be tuned on a dev set. In view of this, we conduct an extensive study of Pet, a method that combines textual instructions with example-based finetuning. We show that, if correctly configured, Pet performs strongly in true few-shot settings without a dev set. Crucial for this strong performance is a number of design choices, including Pet‘s ability to intelligently handle multiple prompts. We put our findings to a real-world test by running Pet on RAFT, a benchmark of tasks taken from realistic NLP applications for which no labeled dev or test sets are available. Pet achieves a new state of the art on RAFT and performs close to non-expert humans for 7 out of 11 tasks. These results demonstrate that prompt-based learners can successfully be applied in true few-shot settings and underpin our belief that learning from instructions will play an important role on the path towards human-like few-shot learning capabilities.
%R 10.1162/tacl_a_00485
%U https://aclanthology.org/2022.tacl-1.41/
%U https://doi.org/10.1162/tacl_a_00485
%P 716-731

Markdown (Informal)

[True Few-Shot Learning with Prompts—A Real-World Perspective](https://aclanthology.org/2022.tacl-1.41/) (Schick & Schütze, TACL 2022)

• True Few-Shot Learning with Prompts—A Real-World Perspective (Schick & Schütze, TACL 2022)

ACL

• Timo Schick and Hinrich Schütze. 2022. True Few-Shot Learning with Prompts—A Real-World Perspective. Transactions of the Association for Computational Linguistics, 10:716–731.