@inproceedings{guo-etal-2026-two,
title = "Two Streams, One Sarcasm: Orthogonal Expert Tuning for Holistic Multimodal Sarcasm Understanding",
author = "Guo, Diandian and
Cao, Cong and
Yuan, Fangfang and
Xu, Pin and
Hu, Cheng and
Zhang, Zhicheng and
Liu, Yu and
Liu, Yanbing",
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.319/",
pages = "7054--7072",
ISBN = "979-8-89176-390-6",
abstract = "Multimodal Sarcasm Understanding (MSU) comprises multiple subtasks, demanding both incongruity perception and intent reasoning. However, this progress is impeded by two bottlenecks. First, the lack of a unified benchmark for holistic satirical cognition hinders comprehensive evaluation of MSU. Second, jointly modeling these heterogeneous subtasks often leads to feature entanglement. Specifically, while subtasks share a dependence on incongruity, they diverge in granular focus, causing specific execution patterns to erode the fundamental perception capability. To address these challenges, we make two contributions. First, we introduce DocMSU-PLUS, a comprehensive benchmark covering five cognitive dimensions of MSU. All tasks are reformulated into multiple-choice questions (MCQs), enabling a unified accuracy-based evaluation. Second, we propose the Dual Orthogonal Stream Experts (DOSE) framework. DOSE structurally decouples experts into orthogonal shared perception and private execution streams to physically block gradient interference between tasks. Experiments demonstrate that DOSE achieves superior performance on DocMSU-PLUS, effectively balancing general perception with task-specific adaptation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="guo-etal-2026-two">
<titleInfo>
<title>Two Streams, One Sarcasm: Orthogonal Expert Tuning for Holistic Multimodal Sarcasm Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Diandian</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cong</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fangfang</namePart>
<namePart type="family">Yuan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pin</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cheng</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhicheng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanbing</namePart>
<namePart type="family">Liu</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>Multimodal Sarcasm Understanding (MSU) comprises multiple subtasks, demanding both incongruity perception and intent reasoning. However, this progress is impeded by two bottlenecks. First, the lack of a unified benchmark for holistic satirical cognition hinders comprehensive evaluation of MSU. Second, jointly modeling these heterogeneous subtasks often leads to feature entanglement. Specifically, while subtasks share a dependence on incongruity, they diverge in granular focus, causing specific execution patterns to erode the fundamental perception capability. To address these challenges, we make two contributions. First, we introduce DocMSU-PLUS, a comprehensive benchmark covering five cognitive dimensions of MSU. All tasks are reformulated into multiple-choice questions (MCQs), enabling a unified accuracy-based evaluation. Second, we propose the Dual Orthogonal Stream Experts (DOSE) framework. DOSE structurally decouples experts into orthogonal shared perception and private execution streams to physically block gradient interference between tasks. Experiments demonstrate that DOSE achieves superior performance on DocMSU-PLUS, effectively balancing general perception with task-specific adaptation.</abstract>
<identifier type="citekey">guo-etal-2026-two</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.319/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>7054</start>
<end>7072</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Two Streams, One Sarcasm: Orthogonal Expert Tuning for Holistic Multimodal Sarcasm Understanding
%A Guo, Diandian
%A Cao, Cong
%A Yuan, Fangfang
%A Xu, Pin
%A Hu, Cheng
%A Zhang, Zhicheng
%A Liu, Yu
%A Liu, Yanbing
%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 guo-etal-2026-two
%X Multimodal Sarcasm Understanding (MSU) comprises multiple subtasks, demanding both incongruity perception and intent reasoning. However, this progress is impeded by two bottlenecks. First, the lack of a unified benchmark for holistic satirical cognition hinders comprehensive evaluation of MSU. Second, jointly modeling these heterogeneous subtasks often leads to feature entanglement. Specifically, while subtasks share a dependence on incongruity, they diverge in granular focus, causing specific execution patterns to erode the fundamental perception capability. To address these challenges, we make two contributions. First, we introduce DocMSU-PLUS, a comprehensive benchmark covering five cognitive dimensions of MSU. All tasks are reformulated into multiple-choice questions (MCQs), enabling a unified accuracy-based evaluation. Second, we propose the Dual Orthogonal Stream Experts (DOSE) framework. DOSE structurally decouples experts into orthogonal shared perception and private execution streams to physically block gradient interference between tasks. Experiments demonstrate that DOSE achieves superior performance on DocMSU-PLUS, effectively balancing general perception with task-specific adaptation.
%U https://aclanthology.org/2026.acl-long.319/
%P 7054-7072
Markdown (Informal)
[Two Streams, One Sarcasm: Orthogonal Expert Tuning for Holistic Multimodal Sarcasm Understanding](https://aclanthology.org/2026.acl-long.319/) (Guo et al., ACL 2026)
ACL
- Diandian Guo, Cong Cao, Fangfang Yuan, Pin Xu, Cheng Hu, Zhicheng Zhang, Yu Liu, and Yanbing Liu. 2026. Two Streams, One Sarcasm: Orthogonal Expert Tuning for Holistic Multimodal Sarcasm Understanding. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7054–7072, San Diego, California, United States. Association for Computational Linguistics.