@inproceedings{shi-etal-2023-toward,
title = "Toward Human Readable Prompt Tuning: Kubrick{'}s The Shining is a good movie, and a good prompt too?",
author = "Shi, Weijia and
Han, Xiaochuang and
Gonen, Hila and
Holtzman, Ari and
Tsvetkov, Yulia and
Zettlemoyer, Luke",
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.733",
doi = "10.18653/v1/2023.findings-emnlp.733",
pages = "10994--11005",
abstract = "Large language models can perform downstream tasks in a zero-shot fashion, given natural language prompts that specify the desired behavior. Such prompts are typically hand engineered, but can also be learned with gradient-based methods from labeled data. However, it is underexplored what factors make the prompts effective, especially when the prompts are in natural language. In this paper, we investigate common attributes shared by effective prompts in classification problems. We first propose a human readable prompt tuning method (FluentPrompt) based on Langevin dynamics that incorporates a fluency constraint to find a distribution of effective and fluent prompts. Our analysis reveals that effective prompts are topically related to the task domain and calibrate the prior probability of output labels. Based on these findings, we also propose a method for generating prompts using only unlabeled data, outperforming strong baselines by an average of 7.0{\%} accuracy across three tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shi-etal-2023-toward">
<titleInfo>
<title>Toward Human Readable Prompt Tuning: Kubrick’s The Shining is a good movie, and a good prompt too?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Weijia</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaochuang</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hila</namePart>
<namePart type="family">Gonen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ari</namePart>
<namePart type="family">Holtzman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulia</namePart>
<namePart type="family">Tsvetkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luke</namePart>
<namePart type="family">Zettlemoyer</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>Large language models can perform downstream tasks in a zero-shot fashion, given natural language prompts that specify the desired behavior. Such prompts are typically hand engineered, but can also be learned with gradient-based methods from labeled data. However, it is underexplored what factors make the prompts effective, especially when the prompts are in natural language. In this paper, we investigate common attributes shared by effective prompts in classification problems. We first propose a human readable prompt tuning method (FluentPrompt) based on Langevin dynamics that incorporates a fluency constraint to find a distribution of effective and fluent prompts. Our analysis reveals that effective prompts are topically related to the task domain and calibrate the prior probability of output labels. Based on these findings, we also propose a method for generating prompts using only unlabeled data, outperforming strong baselines by an average of 7.0% accuracy across three tasks.</abstract>
<identifier type="citekey">shi-etal-2023-toward</identifier>
<identifier type="doi">10.18653/v1/2023.findings-emnlp.733</identifier>
<location>
<url>https://aclanthology.org/2023.findings-emnlp.733</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>10994</start>
<end>11005</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Toward Human Readable Prompt Tuning: Kubrick’s The Shining is a good movie, and a good prompt too?
%A Shi, Weijia
%A Han, Xiaochuang
%A Gonen, Hila
%A Holtzman, Ari
%A Tsvetkov, Yulia
%A Zettlemoyer, Luke
%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 shi-etal-2023-toward
%X Large language models can perform downstream tasks in a zero-shot fashion, given natural language prompts that specify the desired behavior. Such prompts are typically hand engineered, but can also be learned with gradient-based methods from labeled data. However, it is underexplored what factors make the prompts effective, especially when the prompts are in natural language. In this paper, we investigate common attributes shared by effective prompts in classification problems. We first propose a human readable prompt tuning method (FluentPrompt) based on Langevin dynamics that incorporates a fluency constraint to find a distribution of effective and fluent prompts. Our analysis reveals that effective prompts are topically related to the task domain and calibrate the prior probability of output labels. Based on these findings, we also propose a method for generating prompts using only unlabeled data, outperforming strong baselines by an average of 7.0% accuracy across three tasks.
%R 10.18653/v1/2023.findings-emnlp.733
%U https://aclanthology.org/2023.findings-emnlp.733
%U https://doi.org/10.18653/v1/2023.findings-emnlp.733
%P 10994-11005
Markdown (Informal)
[Toward Human Readable Prompt Tuning: Kubrick’s The Shining is a good movie, and a good prompt too?](https://aclanthology.org/2023.findings-emnlp.733) (Shi et al., Findings 2023)
ACL