@inproceedings{mitra-etal-2026-dual,
title = "Dual-Axis Compositional Contrastive Few-Shot Learning using Prototypes Across Linguistic and Semantic Dimensions for {I}ndic Low-Resource Multilingual {NLU}",
author = "Mitra, Kathakali and
Singh, Sakshi and
Gunapati, Sree Nithish Reddy and
Malapati, Aruna and
Lee, Mark G.",
editor = "Chakravarthi, Bharathi Raja and
B, Bharathi and
Buitelaar, Paul and
Thenmozhi, Durairaj and
Garc{\'i}a Cumbreras, Miguel {\'A}ngel and
Jim{\'e}nez Zafra, Salud Mar{\'i}a",
booktitle = "Proceedings of the Sixth Workshop on Language Technology for Equality, Diversity, Inclusion",
month = jul,
year = "2026",
address = "Virtual (Online)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.ltedi-1.3/",
pages = "27--36",
ISBN = "979-8-89176-424-8",
abstract = "Multilingual Natural Language Understanding (NLU) systems often struggle to adapt when new languages or new semantic labels are introduced with only a few annotated examples. This challenge is particularly pronounced for low-resource languages, where limited supervision and evolving label spaces make conventional joint-label classification approaches unstable. Most existing multilingual NLU models treat each language-semantic pair as an independent class, entangling linguistic and semantic representations and hindering few-shot adaptation. We propose Dual-Axis Compositional Few-Shot Learning, a framework that explicitly factorizes the representation space into linguistic and semantic embedding axes, enabling independent modeling of language variation and domain-intent semantics. Joint representations are constructed compositionally through multiplicative interaction of axis-specific embeddings, allowing controlled adaptation when either the language set or the semantic label space evolves. The framework integrates factorized prototype learning, axis-structured contrastive alignment, and disentanglement regularization using HSIC-based statistical independence and Jacobian-based cross-axis decorrelation. Experiments on six low-resource Indic languages spanning Indo-Aryan and Dravidian families (Hindi, Bengali, Sanskrit, Assamese, Tamil, and Telugu) demonstrate strong performance under two structured generalization regimes. The model achieves 81.12{\%} accuracy when adapting to few-shot languages with known semantics and 63.5{\%} accuracy when learning new semantic classes from few-shot examples, along with an accuracy of 89.56{\%} on known language and seen semantics. These results show that axis-factorized representations enable stable compositional generalization, offering a promising direction for scalable multilingual NLU in linguistically diverse low-resource settings."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mitra-etal-2026-dual">
<titleInfo>
<title>Dual-Axis Compositional Contrastive Few-Shot Learning using Prototypes Across Linguistic and Semantic Dimensions for Indic Low-Resource Multilingual NLU</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kathakali</namePart>
<namePart type="family">Mitra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakshi</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sree</namePart>
<namePart type="given">Nithish</namePart>
<namePart type="given">Reddy</namePart>
<namePart type="family">Gunapati</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aruna</namePart>
<namePart type="family">Malapati</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mark</namePart>
<namePart type="given">G</namePart>
<namePart type="family">Lee</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 Sixth Workshop on Language Technology for Equality, Diversity, Inclusion</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bharathi</namePart>
<namePart type="given">Raja</namePart>
<namePart type="family">Chakravarthi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bharathi</namePart>
<namePart type="family">B</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">Buitelaar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Durairaj</namePart>
<namePart type="family">Thenmozhi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miguel</namePart>
<namePart type="given">Ángel</namePart>
<namePart type="family">García Cumbreras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salud</namePart>
<namePart type="given">María</namePart>
<namePart type="family">Jiménez Zafra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Virtual (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-424-8</identifier>
</relatedItem>
<abstract>Multilingual Natural Language Understanding (NLU) systems often struggle to adapt when new languages or new semantic labels are introduced with only a few annotated examples. This challenge is particularly pronounced for low-resource languages, where limited supervision and evolving label spaces make conventional joint-label classification approaches unstable. Most existing multilingual NLU models treat each language-semantic pair as an independent class, entangling linguistic and semantic representations and hindering few-shot adaptation. We propose Dual-Axis Compositional Few-Shot Learning, a framework that explicitly factorizes the representation space into linguistic and semantic embedding axes, enabling independent modeling of language variation and domain-intent semantics. Joint representations are constructed compositionally through multiplicative interaction of axis-specific embeddings, allowing controlled adaptation when either the language set or the semantic label space evolves. The framework integrates factorized prototype learning, axis-structured contrastive alignment, and disentanglement regularization using HSIC-based statistical independence and Jacobian-based cross-axis decorrelation. Experiments on six low-resource Indic languages spanning Indo-Aryan and Dravidian families (Hindi, Bengali, Sanskrit, Assamese, Tamil, and Telugu) demonstrate strong performance under two structured generalization regimes. The model achieves 81.12% accuracy when adapting to few-shot languages with known semantics and 63.5% accuracy when learning new semantic classes from few-shot examples, along with an accuracy of 89.56% on known language and seen semantics. These results show that axis-factorized representations enable stable compositional generalization, offering a promising direction for scalable multilingual NLU in linguistically diverse low-resource settings.</abstract>
<identifier type="citekey">mitra-etal-2026-dual</identifier>
<location>
<url>https://aclanthology.org/2026.ltedi-1.3/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>27</start>
<end>36</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Dual-Axis Compositional Contrastive Few-Shot Learning using Prototypes Across Linguistic and Semantic Dimensions for Indic Low-Resource Multilingual NLU
%A Mitra, Kathakali
%A Singh, Sakshi
%A Gunapati, Sree Nithish Reddy
%A Malapati, Aruna
%A Lee, Mark G.
%Y Chakravarthi, Bharathi Raja
%Y B, Bharathi
%Y Buitelaar, Paul
%Y Thenmozhi, Durairaj
%Y García Cumbreras, Miguel Ángel
%Y Jiménez Zafra, Salud María
%S Proceedings of the Sixth Workshop on Language Technology for Equality, Diversity, Inclusion
%D 2026
%8 July
%I Association for Computational Linguistics
%C Virtual (Online)
%@ 979-8-89176-424-8
%F mitra-etal-2026-dual
%X Multilingual Natural Language Understanding (NLU) systems often struggle to adapt when new languages or new semantic labels are introduced with only a few annotated examples. This challenge is particularly pronounced for low-resource languages, where limited supervision and evolving label spaces make conventional joint-label classification approaches unstable. Most existing multilingual NLU models treat each language-semantic pair as an independent class, entangling linguistic and semantic representations and hindering few-shot adaptation. We propose Dual-Axis Compositional Few-Shot Learning, a framework that explicitly factorizes the representation space into linguistic and semantic embedding axes, enabling independent modeling of language variation and domain-intent semantics. Joint representations are constructed compositionally through multiplicative interaction of axis-specific embeddings, allowing controlled adaptation when either the language set or the semantic label space evolves. The framework integrates factorized prototype learning, axis-structured contrastive alignment, and disentanglement regularization using HSIC-based statistical independence and Jacobian-based cross-axis decorrelation. Experiments on six low-resource Indic languages spanning Indo-Aryan and Dravidian families (Hindi, Bengali, Sanskrit, Assamese, Tamil, and Telugu) demonstrate strong performance under two structured generalization regimes. The model achieves 81.12% accuracy when adapting to few-shot languages with known semantics and 63.5% accuracy when learning new semantic classes from few-shot examples, along with an accuracy of 89.56% on known language and seen semantics. These results show that axis-factorized representations enable stable compositional generalization, offering a promising direction for scalable multilingual NLU in linguistically diverse low-resource settings.
%U https://aclanthology.org/2026.ltedi-1.3/
%P 27-36
Markdown (Informal)
[Dual-Axis Compositional Contrastive Few-Shot Learning using Prototypes Across Linguistic and Semantic Dimensions for Indic Low-Resource Multilingual NLU](https://aclanthology.org/2026.ltedi-1.3/) (Mitra et al., LTEDI 2026)
ACL