@inproceedings{dao-sy-etal-2026-hcmus-fangs,
title = "{HCMUS}{\_}{T}he Fangs at {A}bjad{S}tyle{T}ransfer Shared Task: Learning to Query Style, Contrastive Representations for Zero-Shot {A}rabic Authorship Style Transfer",
author = "Dao Sy, Duy Minh and
Huynh, Trung Kiet and
Tran, Nguyen Chi and
Quy, Nguyen Lam Phu and
Hoa, Pham Phu and
Duong, Nguyen Dinh Ha",
booktitle = "Proceedings of the 2nd Workshop on {NLP} for Languages Using {A}rabic Script",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.abjadnlp-1.52/",
pages = "438--442",
abstract = "This paper describes the system developed by team HCMUS{\_}The Fangs for the AbjadStyleTransfer shared task (ArabicNLP 2026), where we achieved 1st place. We present a contrastive style learning approach for zero-shot Arabic authorship style transfer. Our key discovery is that the 21 test authors-including Nobel laureate Naguib Mahfouz and literary pioneer Taha Hussein-have zero overlap with the 32,784 training authors, transforming this into a pure zero-shot challenge. This insight led us to develop a dual-encoder architecture that learns transferable style representations through contrastive objectives, rather than memorizing author-specific patterns. Our system achieves 19.77 BLEU and 55.74 chrF, outperforming retrieval-augmented generation (+18{\%}) and multi-task learning (+31{\%}). Counter-intuitively, we find that sophisticated architectural modifications like style injection consistently degrade performance, while simpler approaches that preserve pre-trained knowledge excel. Our analysis reveals that for famous authors, pre-trained Arabic language models already encode substantial stylistic knowledge-the key is surfacing it, not learning from scratch."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dao-sy-etal-2026-hcmus-fangs">
<titleInfo>
<title>HCMUS_The Fangs at AbjadStyleTransfer Shared Task: Learning to Query Style, Contrastive Representations for Zero-Shot Arabic Authorship Style Transfer</title>
</titleInfo>
<name type="personal">
<namePart type="given">Duy</namePart>
<namePart type="given">Minh</namePart>
<namePart type="family">Dao Sy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Trung</namePart>
<namePart type="given">Kiet</namePart>
<namePart type="family">Huynh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nguyen</namePart>
<namePart type="given">Chi</namePart>
<namePart type="family">Tran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nguyen</namePart>
<namePart type="given">Lam</namePart>
<namePart type="given">Phu</namePart>
<namePart type="family">Quy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pham</namePart>
<namePart type="given">Phu</namePart>
<namePart type="family">Hoa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nguyen</namePart>
<namePart type="given">Dinh</namePart>
<namePart type="given">Ha</namePart>
<namePart type="family">Duong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes the system developed by team HCMUS_The Fangs for the AbjadStyleTransfer shared task (ArabicNLP 2026), where we achieved 1st place. We present a contrastive style learning approach for zero-shot Arabic authorship style transfer. Our key discovery is that the 21 test authors-including Nobel laureate Naguib Mahfouz and literary pioneer Taha Hussein-have zero overlap with the 32,784 training authors, transforming this into a pure zero-shot challenge. This insight led us to develop a dual-encoder architecture that learns transferable style representations through contrastive objectives, rather than memorizing author-specific patterns. Our system achieves 19.77 BLEU and 55.74 chrF, outperforming retrieval-augmented generation (+18%) and multi-task learning (+31%). Counter-intuitively, we find that sophisticated architectural modifications like style injection consistently degrade performance, while simpler approaches that preserve pre-trained knowledge excel. Our analysis reveals that for famous authors, pre-trained Arabic language models already encode substantial stylistic knowledge-the key is surfacing it, not learning from scratch.</abstract>
<identifier type="citekey">dao-sy-etal-2026-hcmus-fangs</identifier>
<location>
<url>https://aclanthology.org/2026.abjadnlp-1.52/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>438</start>
<end>442</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HCMUS_The Fangs at AbjadStyleTransfer Shared Task: Learning to Query Style, Contrastive Representations for Zero-Shot Arabic Authorship Style Transfer
%A Dao Sy, Duy Minh
%A Huynh, Trung Kiet
%A Tran, Nguyen Chi
%A Quy, Nguyen Lam Phu
%A Hoa, Pham Phu
%A Duong, Nguyen Dinh Ha
%S Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%F dao-sy-etal-2026-hcmus-fangs
%X This paper describes the system developed by team HCMUS_The Fangs for the AbjadStyleTransfer shared task (ArabicNLP 2026), where we achieved 1st place. We present a contrastive style learning approach for zero-shot Arabic authorship style transfer. Our key discovery is that the 21 test authors-including Nobel laureate Naguib Mahfouz and literary pioneer Taha Hussein-have zero overlap with the 32,784 training authors, transforming this into a pure zero-shot challenge. This insight led us to develop a dual-encoder architecture that learns transferable style representations through contrastive objectives, rather than memorizing author-specific patterns. Our system achieves 19.77 BLEU and 55.74 chrF, outperforming retrieval-augmented generation (+18%) and multi-task learning (+31%). Counter-intuitively, we find that sophisticated architectural modifications like style injection consistently degrade performance, while simpler approaches that preserve pre-trained knowledge excel. Our analysis reveals that for famous authors, pre-trained Arabic language models already encode substantial stylistic knowledge-the key is surfacing it, not learning from scratch.
%U https://aclanthology.org/2026.abjadnlp-1.52/
%P 438-442
Markdown (Informal)
[HCMUS_The Fangs at AbjadStyleTransfer Shared Task: Learning to Query Style, Contrastive Representations for Zero-Shot Arabic Authorship Style Transfer](https://aclanthology.org/2026.abjadnlp-1.52/) (Dao Sy et al., AbjadNLP 2026)
ACL