@inproceedings{ashok-etal-2023-student,
title = "The student becomes the master: Outperforming {GPT}3 on Scientific Factual Error Correction",
author = "Ashok, Dhananjay and
Kulkarni, Atharva and
Pham, Hai and
Poczos, Barnabas",
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.451",
doi = "10.18653/v1/2023.findings-emnlp.451",
pages = "6762--6778",
abstract = "Due to the prohibitively high cost of creating error correction datasets, most Factual Claim Correction methods rely on a powerful verification model to guide the correction process. This leads to a significant drop in performance in domains like Scientific Claim Correction, where good verification models do not always exist. In this work we introduce SciFix, a claim correction system that does not require a verifier but is able to outperform existing methods by a considerable margin {---} achieving correction accuracy of 84{\%} on the SciFact dataset, 77{\%} on SciFact-Open and 72.75{\%} on the CovidFact dataset, compared to next best accuracies of 7.6{\%}, 5{\%} and 15{\%} on the same datasets respectively. Our method leverages the power of prompting with LLMs during training to create a richly annotated dataset that can be used for fully supervised training and regularization. We additionally use a claim-aware decoding procedure to improve the quality of corrected claims. Our method outperforms the very LLM that was used to generate the annotated dataset {---} with FewShot Prompting on GPT3.5 achieving 58{\%}, 61{\%} and 64{\%} on the respective datasets, a consistently lower correction accuracy, despite using nearly 800 times as many parameters as our model.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ashok-etal-2023-student">
<titleInfo>
<title>The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dhananjay</namePart>
<namePart type="family">Ashok</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Atharva</namePart>
<namePart type="family">Kulkarni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hai</namePart>
<namePart type="family">Pham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barnabas</namePart>
<namePart type="family">Poczos</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>Due to the prohibitively high cost of creating error correction datasets, most Factual Claim Correction methods rely on a powerful verification model to guide the correction process. This leads to a significant drop in performance in domains like Scientific Claim Correction, where good verification models do not always exist. In this work we introduce SciFix, a claim correction system that does not require a verifier but is able to outperform existing methods by a considerable margin — achieving correction accuracy of 84% on the SciFact dataset, 77% on SciFact-Open and 72.75% on the CovidFact dataset, compared to next best accuracies of 7.6%, 5% and 15% on the same datasets respectively. Our method leverages the power of prompting with LLMs during training to create a richly annotated dataset that can be used for fully supervised training and regularization. We additionally use a claim-aware decoding procedure to improve the quality of corrected claims. Our method outperforms the very LLM that was used to generate the annotated dataset — with FewShot Prompting on GPT3.5 achieving 58%, 61% and 64% on the respective datasets, a consistently lower correction accuracy, despite using nearly 800 times as many parameters as our model.</abstract>
<identifier type="citekey">ashok-etal-2023-student</identifier>
<identifier type="doi">10.18653/v1/2023.findings-emnlp.451</identifier>
<location>
<url>https://aclanthology.org/2023.findings-emnlp.451</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>6762</start>
<end>6778</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction
%A Ashok, Dhananjay
%A Kulkarni, Atharva
%A Pham, Hai
%A Poczos, Barnabas
%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 ashok-etal-2023-student
%X Due to the prohibitively high cost of creating error correction datasets, most Factual Claim Correction methods rely on a powerful verification model to guide the correction process. This leads to a significant drop in performance in domains like Scientific Claim Correction, where good verification models do not always exist. In this work we introduce SciFix, a claim correction system that does not require a verifier but is able to outperform existing methods by a considerable margin — achieving correction accuracy of 84% on the SciFact dataset, 77% on SciFact-Open and 72.75% on the CovidFact dataset, compared to next best accuracies of 7.6%, 5% and 15% on the same datasets respectively. Our method leverages the power of prompting with LLMs during training to create a richly annotated dataset that can be used for fully supervised training and regularization. We additionally use a claim-aware decoding procedure to improve the quality of corrected claims. Our method outperforms the very LLM that was used to generate the annotated dataset — with FewShot Prompting on GPT3.5 achieving 58%, 61% and 64% on the respective datasets, a consistently lower correction accuracy, despite using nearly 800 times as many parameters as our model.
%R 10.18653/v1/2023.findings-emnlp.451
%U https://aclanthology.org/2023.findings-emnlp.451
%U https://doi.org/10.18653/v1/2023.findings-emnlp.451
%P 6762-6778
Markdown (Informal)
[The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction](https://aclanthology.org/2023.findings-emnlp.451) (Ashok et al., Findings 2023)
ACL