@inproceedings{liu-etal-2026-llms-hear,
title = "Can {LLM}s Hear the Dogwhistle?",
author = "Liu, Yifan and
Lin, Yi and
Guo, Xinwei and
Wang, Ziwei and
Zhang, Jiaxin and
Chen, Guanhua and
Wu, Haiyan and
Zhao, Xiangyu and
Yao, Xin and
Wei, Xuetao",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.161/",
pages = "3256--3273",
ISBN = "979-8-89176-395-1",
abstract = "With the widespread deployment of large language models (LLMs), existing safety benchmarks remain largely focused on explicitly harmful content, overlooking context-dependent expressions such as dogwhistles, the language that conveys harmful intent while appearing benign on the surface. To address this gap, we introduce DogBench, a comprehensive benchmark for evaluating LLM safety under dogwhistle-driven prompts. DogBench comprises 11,150 prompt instances constructed from controlled templates that embed dogwhistle terms, allowing for enabling direct comparison with explicit toxic terms under identical prompt structures. Each prompt is further annotated with pragmatic attributes, including interaction category and stance tendency. Extensive evaluations across multiple mainstream LLMs reveal a consistent pattern: dogwhistle prompts are substantially more likely to elicit harmful outputs than their explicit toxic counterparts, with an average risk increase of approximately fourfold. These findings expose a blind spot in current safety evaluation and alignment practices. Our work underscores the need to explicitly incorporate dogwhistles into future LLM safety research, with DogBench serving as a dedicated benchmark for this purpose."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-llms-hear">
<titleInfo>
<title>Can LLMs Hear the Dogwhistle?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yifan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinwei</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ziwei</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaxin</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guanhua</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haiyan</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangyu</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Yao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuetao</namePart>
<namePart type="family">Wei</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>Findings of the Association for Computational Linguistics: ACL 2026</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">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>With the widespread deployment of large language models (LLMs), existing safety benchmarks remain largely focused on explicitly harmful content, overlooking context-dependent expressions such as dogwhistles, the language that conveys harmful intent while appearing benign on the surface. To address this gap, we introduce DogBench, a comprehensive benchmark for evaluating LLM safety under dogwhistle-driven prompts. DogBench comprises 11,150 prompt instances constructed from controlled templates that embed dogwhistle terms, allowing for enabling direct comparison with explicit toxic terms under identical prompt structures. Each prompt is further annotated with pragmatic attributes, including interaction category and stance tendency. Extensive evaluations across multiple mainstream LLMs reveal a consistent pattern: dogwhistle prompts are substantially more likely to elicit harmful outputs than their explicit toxic counterparts, with an average risk increase of approximately fourfold. These findings expose a blind spot in current safety evaluation and alignment practices. Our work underscores the need to explicitly incorporate dogwhistles into future LLM safety research, with DogBench serving as a dedicated benchmark for this purpose.</abstract>
<identifier type="citekey">liu-etal-2026-llms-hear</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.161/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>3256</start>
<end>3273</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Can LLMs Hear the Dogwhistle?
%A Liu, Yifan
%A Lin, Yi
%A Guo, Xinwei
%A Wang, Ziwei
%A Zhang, Jiaxin
%A Chen, Guanhua
%A Wu, Haiyan
%A Zhao, Xiangyu
%A Yao, Xin
%A Wei, Xuetao
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F liu-etal-2026-llms-hear
%X With the widespread deployment of large language models (LLMs), existing safety benchmarks remain largely focused on explicitly harmful content, overlooking context-dependent expressions such as dogwhistles, the language that conveys harmful intent while appearing benign on the surface. To address this gap, we introduce DogBench, a comprehensive benchmark for evaluating LLM safety under dogwhistle-driven prompts. DogBench comprises 11,150 prompt instances constructed from controlled templates that embed dogwhistle terms, allowing for enabling direct comparison with explicit toxic terms under identical prompt structures. Each prompt is further annotated with pragmatic attributes, including interaction category and stance tendency. Extensive evaluations across multiple mainstream LLMs reveal a consistent pattern: dogwhistle prompts are substantially more likely to elicit harmful outputs than their explicit toxic counterparts, with an average risk increase of approximately fourfold. These findings expose a blind spot in current safety evaluation and alignment practices. Our work underscores the need to explicitly incorporate dogwhistles into future LLM safety research, with DogBench serving as a dedicated benchmark for this purpose.
%U https://aclanthology.org/2026.findings-acl.161/
%P 3256-3273
Markdown (Informal)
[Can LLMs Hear the Dogwhistle?](https://aclanthology.org/2026.findings-acl.161/) (Liu et al., Findings 2026)
ACL
- Yifan Liu, Yi Lin, Xinwei Guo, Ziwei Wang, Jiaxin Zhang, Guanhua Chen, Haiyan Wu, Xiangyu Zhao, Xin Yao, and Xuetao Wei. 2026. Can LLMs Hear the Dogwhistle?. In Findings of the Association for Computational Linguistics: ACL 2026, pages 3256–3273, San Diego, California, United States. Association for Computational Linguistics.