@inproceedings{islam-surdeanu-2026-lightweight,
title = "A Lightweight Explainable Guardrail for Prompt Safety",
author = "Islam, Md Asiful and
Surdeanu, Mihai",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.2017/",
doi = "10.18653/v1/2026.acl-long.2017",
pages = "43573--43591",
ISBN = "979-8-89176-390-6",
abstract = "We propose a lightweight explainable guardrail (LEG) method to detect unsafe prompts. LEG uses a multi-task learning architecture to jointly learn a prompt classifier and an explanation classifier, where the latter labels prompt words that explain the safe/unsafe overall decision. LEG is trained on synthetic explanation data, which is generated using a novel strategy that counteracts the confirmation biases of LLMs. Lastly, LEG{'}s training process uses a novel loss that captures global explanation signals as a weak supervision and combines cross-entropy and focal losses with uncertainty-based weighting. LEG obtains equivalent or better performance than the state-of-the-art for both prompt classification and explainability, both in-domain and out-of-domain on three datasets, despite the fact that its model size is considerably smaller than current approaches."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="islam-surdeanu-2026-lightweight">
<titleInfo>
<title>A Lightweight Explainable Guardrail for Prompt Safety</title>
</titleInfo>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="given">Asiful</namePart>
<namePart type="family">Islam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mihai</namePart>
<namePart type="family">Surdeanu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>We propose a lightweight explainable guardrail (LEG) method to detect unsafe prompts. LEG uses a multi-task learning architecture to jointly learn a prompt classifier and an explanation classifier, where the latter labels prompt words that explain the safe/unsafe overall decision. LEG is trained on synthetic explanation data, which is generated using a novel strategy that counteracts the confirmation biases of LLMs. Lastly, LEG’s training process uses a novel loss that captures global explanation signals as a weak supervision and combines cross-entropy and focal losses with uncertainty-based weighting. LEG obtains equivalent or better performance than the state-of-the-art for both prompt classification and explainability, both in-domain and out-of-domain on three datasets, despite the fact that its model size is considerably smaller than current approaches.</abstract>
<identifier type="citekey">islam-surdeanu-2026-lightweight</identifier>
<identifier type="doi">10.18653/v1/2026.acl-long.2017</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.2017/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>43573</start>
<end>43591</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Lightweight Explainable Guardrail for Prompt Safety
%A Islam, Md Asiful
%A Surdeanu, Mihai
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F islam-surdeanu-2026-lightweight
%X We propose a lightweight explainable guardrail (LEG) method to detect unsafe prompts. LEG uses a multi-task learning architecture to jointly learn a prompt classifier and an explanation classifier, where the latter labels prompt words that explain the safe/unsafe overall decision. LEG is trained on synthetic explanation data, which is generated using a novel strategy that counteracts the confirmation biases of LLMs. Lastly, LEG’s training process uses a novel loss that captures global explanation signals as a weak supervision and combines cross-entropy and focal losses with uncertainty-based weighting. LEG obtains equivalent or better performance than the state-of-the-art for both prompt classification and explainability, both in-domain and out-of-domain on three datasets, despite the fact that its model size is considerably smaller than current approaches.
%R 10.18653/v1/2026.acl-long.2017
%U https://aclanthology.org/2026.acl-long.2017/
%U https://doi.org/10.18653/v1/2026.acl-long.2017
%P 43573-43591
Markdown (Informal)
[A Lightweight Explainable Guardrail for Prompt Safety](https://aclanthology.org/2026.acl-long.2017/) (Islam & Surdeanu, ACL 2026)
ACL
- Md Asiful Islam and Mihai Surdeanu. 2026. A Lightweight Explainable Guardrail for Prompt Safety. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 43573–43591, San Diego, California, United States. Association for Computational Linguistics.