@inproceedings{shen-etal-2023-flatness,
title = "Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency",
author = "Shen, Lingfeng and
Tan, Weiting and
Zheng, Boyuan and
Khashabi, Daniel",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2023",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-emnlp.523",
doi = "10.18653/v1/2023.findings-emnlp.523",
pages = "7795--7817",
abstract = "With growing capabilities of large language models, prompting them has become the dominant way to access them. This has motivated the development of strategies for automatically selecting effective language prompts. In this paper, we introduce **pFlat** (prompt flatness), a new metric to quantify the expected utility of a language prompt. This metric is inspired by *flatness* regularization in statistical learning that quantifies the robustness of the model towards its parameter perturbations. We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods. Empirically, we show that combining **pFlat** with existing metrics improves both performance and sample efficiency. Our metric outperforms the previous prompt selection metrics with an average increase of 10{\%} in Pearson correlation across 6 classification benchmarks, and the prompt selected by our metric gains 5{\%} higher accuracy than previous metrics across the benchmarks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shen-etal-2023-flatness">
<titleInfo>
<title>Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lingfeng</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weiting</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Boyuan</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Khashabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="family">Pino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>With growing capabilities of large language models, prompting them has become the dominant way to access them. This has motivated the development of strategies for automatically selecting effective language prompts. In this paper, we introduce **pFlat** (prompt flatness), a new metric to quantify the expected utility of a language prompt. This metric is inspired by *flatness* regularization in statistical learning that quantifies the robustness of the model towards its parameter perturbations. We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods. Empirically, we show that combining **pFlat** with existing metrics improves both performance and sample efficiency. Our metric outperforms the previous prompt selection metrics with an average increase of 10% in Pearson correlation across 6 classification benchmarks, and the prompt selected by our metric gains 5% higher accuracy than previous metrics across the benchmarks.</abstract>
<identifier type="citekey">shen-etal-2023-flatness</identifier>
<identifier type="doi">10.18653/v1/2023.findings-emnlp.523</identifier>
<location>
<url>https://aclanthology.org/2023.findings-emnlp.523</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>7795</start>
<end>7817</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency
%A Shen, Lingfeng
%A Tan, Weiting
%A Zheng, Boyuan
%A Khashabi, Daniel
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Findings of the Association for Computational Linguistics: EMNLP 2023
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F shen-etal-2023-flatness
%X With growing capabilities of large language models, prompting them has become the dominant way to access them. This has motivated the development of strategies for automatically selecting effective language prompts. In this paper, we introduce **pFlat** (prompt flatness), a new metric to quantify the expected utility of a language prompt. This metric is inspired by *flatness* regularization in statistical learning that quantifies the robustness of the model towards its parameter perturbations. We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods. Empirically, we show that combining **pFlat** with existing metrics improves both performance and sample efficiency. Our metric outperforms the previous prompt selection metrics with an average increase of 10% in Pearson correlation across 6 classification benchmarks, and the prompt selected by our metric gains 5% higher accuracy than previous metrics across the benchmarks.
%R 10.18653/v1/2023.findings-emnlp.523
%U https://aclanthology.org/2023.findings-emnlp.523
%U https://doi.org/10.18653/v1/2023.findings-emnlp.523
%P 7795-7817
Markdown (Informal)
[Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency](https://aclanthology.org/2023.findings-emnlp.523) (Shen et al., Findings 2023)
ACL