@inproceedings{hua-etal-2024-propagation,
title = "Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks",
author = "Hua, Wenyue and
Guo, Jiang and
Dong, Mingwen and
Zhu, Henghui and
Ng, Patrick and
Wang, Zhiguo",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.743",
doi = "10.18653/v1/2024.findings-acl.743",
pages = "12503--12525",
abstract = "Current knowledge editing approaches struggle to effectively propagate updates to interconnected facts.In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasoning. To support our analysis, we introduce a novel reasoning-based benchmark, ReCoE (Reasoning-based Counterfactual Editing dataset), which covers six common reasoning schemes in the real world. We conduct an extensive analysis of existing knowledge editing techniques, including input-augmentation, finetuning, and locate-and-edit methods. We found that all model editing methods exhibit notably low performance on this dataset, especially within certain reasoning schemes. Our analysis of the chain-of-thought responses from edited models indicate that, while the models effectively update individual facts, they struggle to recall these facts in reasoning tasks. Moreover, locate-and-edit methods severely deteriorate the models{'} language modeling capabilities, leading to poor perplexity and logical coherence in their outputs.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hua-etal-2024-propagation">
<titleInfo>
<title>Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wenyue</namePart>
<namePart type="family">Hua</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiang</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingwen</namePart>
<namePart type="family">Dong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henghui</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patrick</namePart>
<namePart type="family">Ng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiguo</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Current knowledge editing approaches struggle to effectively propagate updates to interconnected facts.In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasoning. To support our analysis, we introduce a novel reasoning-based benchmark, ReCoE (Reasoning-based Counterfactual Editing dataset), which covers six common reasoning schemes in the real world. We conduct an extensive analysis of existing knowledge editing techniques, including input-augmentation, finetuning, and locate-and-edit methods. We found that all model editing methods exhibit notably low performance on this dataset, especially within certain reasoning schemes. Our analysis of the chain-of-thought responses from edited models indicate that, while the models effectively update individual facts, they struggle to recall these facts in reasoning tasks. Moreover, locate-and-edit methods severely deteriorate the models’ language modeling capabilities, leading to poor perplexity and logical coherence in their outputs.</abstract>
<identifier type="citekey">hua-etal-2024-propagation</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.743</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.743</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>12503</start>
<end>12525</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks
%A Hua, Wenyue
%A Guo, Jiang
%A Dong, Mingwen
%A Zhu, Henghui
%A Ng, Patrick
%A Wang, Zhiguo
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F hua-etal-2024-propagation
%X Current knowledge editing approaches struggle to effectively propagate updates to interconnected facts.In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasoning. To support our analysis, we introduce a novel reasoning-based benchmark, ReCoE (Reasoning-based Counterfactual Editing dataset), which covers six common reasoning schemes in the real world. We conduct an extensive analysis of existing knowledge editing techniques, including input-augmentation, finetuning, and locate-and-edit methods. We found that all model editing methods exhibit notably low performance on this dataset, especially within certain reasoning schemes. Our analysis of the chain-of-thought responses from edited models indicate that, while the models effectively update individual facts, they struggle to recall these facts in reasoning tasks. Moreover, locate-and-edit methods severely deteriorate the models’ language modeling capabilities, leading to poor perplexity and logical coherence in their outputs.
%R 10.18653/v1/2024.findings-acl.743
%U https://aclanthology.org/2024.findings-acl.743
%U https://doi.org/10.18653/v1/2024.findings-acl.743
%P 12503-12525
Markdown (Informal)
[Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks](https://aclanthology.org/2024.findings-acl.743) (Hua et al., Findings 2024)
ACL