@inproceedings{he-etal-2026-model,
title = "Can Model Editing Survive Fine-Tuning? Evaluating Robustness and Persistence in Text-to-Image Diffusion Models",
author = "He, Feng and
Valentino, Marco and
Zhao, Zhixue",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-aacl.96/",
pages = "1926--1946",
ISBN = "979-8-89176-444-6",
abstract = "Model editing enables targeted behavioral changes in pre-trained models without costly retraining, supporting applications such as factual correction, debiasing, and safety mitigation. In practice, models are often subsequently fine-tuned for downstream tasks, raising the question of whether edits persist or are reversed during fine-tuning. Understanding this interaction is critical for AI safety, as fine-tuning may either remediate malicious edits or inadvertently undo beneficial ones. We systematically study the impact of subsequent fine-tuning on model editing in text-to-image models. Using two model families (Stable Diffusion and FLUX), five editing methods, and four fine-tuning approaches, we evaluate edit retention of multiple editing objectives under fine-tuning on unrelated tasks. Across models, methods, and metrics, we find that fine-tuning moderately weakens appearance and debiasing edits, while somewhat strengthening edits that remove unsafe content, e.g., for appearance editing, fine-tuning reverses avg. 6.78{\%} editing effect across methods. Further, we provide a mechanistic analysis of edit preservation by examining cross-attention representations and disentangling update magnitude from functional behavioral impact, finding that robustness is associated with systematic representation shifts and limited interference with the edited components. These findings highlight the need for monitoring and, when necessary, reapplying edits in model deployment pipelines."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2026-model">
<titleInfo>
<title>Can Model Editing Survive Fine-Tuning? Evaluating Robustness and Persistence in Text-to-Image Diffusion Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Feng</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Valentino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhixue</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-444-6</identifier>
</relatedItem>
<abstract>Model editing enables targeted behavioral changes in pre-trained models without costly retraining, supporting applications such as factual correction, debiasing, and safety mitigation. In practice, models are often subsequently fine-tuned for downstream tasks, raising the question of whether edits persist or are reversed during fine-tuning. Understanding this interaction is critical for AI safety, as fine-tuning may either remediate malicious edits or inadvertently undo beneficial ones. We systematically study the impact of subsequent fine-tuning on model editing in text-to-image models. Using two model families (Stable Diffusion and FLUX), five editing methods, and four fine-tuning approaches, we evaluate edit retention of multiple editing objectives under fine-tuning on unrelated tasks. Across models, methods, and metrics, we find that fine-tuning moderately weakens appearance and debiasing edits, while somewhat strengthening edits that remove unsafe content, e.g., for appearance editing, fine-tuning reverses avg. 6.78% editing effect across methods. Further, we provide a mechanistic analysis of edit preservation by examining cross-attention representations and disentangling update magnitude from functional behavioral impact, finding that robustness is associated with systematic representation shifts and limited interference with the edited components. These findings highlight the need for monitoring and, when necessary, reapplying edits in model deployment pipelines.</abstract>
<identifier type="citekey">he-etal-2026-model</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.96/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>1926</start>
<end>1946</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Can Model Editing Survive Fine-Tuning? Evaluating Robustness and Persistence in Text-to-Image Diffusion Models
%A He, Feng
%A Valentino, Marco
%A Zhao, Zhixue
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F he-etal-2026-model
%X Model editing enables targeted behavioral changes in pre-trained models without costly retraining, supporting applications such as factual correction, debiasing, and safety mitigation. In practice, models are often subsequently fine-tuned for downstream tasks, raising the question of whether edits persist or are reversed during fine-tuning. Understanding this interaction is critical for AI safety, as fine-tuning may either remediate malicious edits or inadvertently undo beneficial ones. We systematically study the impact of subsequent fine-tuning on model editing in text-to-image models. Using two model families (Stable Diffusion and FLUX), five editing methods, and four fine-tuning approaches, we evaluate edit retention of multiple editing objectives under fine-tuning on unrelated tasks. Across models, methods, and metrics, we find that fine-tuning moderately weakens appearance and debiasing edits, while somewhat strengthening edits that remove unsafe content, e.g., for appearance editing, fine-tuning reverses avg. 6.78% editing effect across methods. Further, we provide a mechanistic analysis of edit preservation by examining cross-attention representations and disentangling update magnitude from functional behavioral impact, finding that robustness is associated with systematic representation shifts and limited interference with the edited components. These findings highlight the need for monitoring and, when necessary, reapplying edits in model deployment pipelines.
%U https://aclanthology.org/2026.findings-aacl.96/
%P 1926-1946
Markdown (Informal)
[Can Model Editing Survive Fine-Tuning? Evaluating Robustness and Persistence in Text-to-Image Diffusion Models](https://aclanthology.org/2026.findings-aacl.96/) (He et al., Findings 2026)
ACL