@inproceedings{kumar-toshniwal-2025-semantic,
title = "Semantic alignment in hyperbolic space for fine-grained emotion classification",
author = "Kumar, Ashish and
Toshniwal, Durga",
editor = "Zhao, Jin and
Wang, Mingyang and
Liu, Zhu",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-srw.55/",
doi = "10.18653/v1/2025.acl-srw.55",
pages = "806--813",
ISBN = "979-8-89176-254-1",
abstract = "Existing approaches to fine-grained emotion classification (FEC) often operate in Euclidean space, where the flat geometry limits the ability to distinguish semantically similar emotion labels (e.g., *annoyed* vs. *angry*). While prior research has explored hyperbolic geometry to capture fine-grained label distinctions, it typically relies on predefined hierarchies and ignores semantically similar negative labels that can mislead the model into making incorrect predictions. In this work, we propose HyCoEM (Hyperbolic Contrastive Learning for Emotion Classification), a semantic alignment framework that leverages the Lorentz model of hyperbolic space. Our approach embeds text and label representations into hyperbolic space via the exponential map, and employs a contrastive loss to bring text embeddings closer to their true labels while pushing them away from adaptively selected, semantically similar negatives. This enables the model to learn label embeddings without relying on a predefined hierarchy and better captures subtle distinctions by incorporating information from both positive and challenging negative labels. Experimental results on two benchmark FEC datasets demonstrate the effectiveness of our approach over baseline methods."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kumar-toshniwal-2025-semantic">
<titleInfo>
<title>Semantic alignment in hyperbolic space for fine-grained emotion classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ashish</namePart>
<namePart type="family">Kumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Durga</namePart>
<namePart type="family">Toshniwal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jin</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-254-1</identifier>
</relatedItem>
<abstract>Existing approaches to fine-grained emotion classification (FEC) often operate in Euclidean space, where the flat geometry limits the ability to distinguish semantically similar emotion labels (e.g., *annoyed* vs. *angry*). While prior research has explored hyperbolic geometry to capture fine-grained label distinctions, it typically relies on predefined hierarchies and ignores semantically similar negative labels that can mislead the model into making incorrect predictions. In this work, we propose HyCoEM (Hyperbolic Contrastive Learning for Emotion Classification), a semantic alignment framework that leverages the Lorentz model of hyperbolic space. Our approach embeds text and label representations into hyperbolic space via the exponential map, and employs a contrastive loss to bring text embeddings closer to their true labels while pushing them away from adaptively selected, semantically similar negatives. This enables the model to learn label embeddings without relying on a predefined hierarchy and better captures subtle distinctions by incorporating information from both positive and challenging negative labels. Experimental results on two benchmark FEC datasets demonstrate the effectiveness of our approach over baseline methods.</abstract>
<identifier type="citekey">kumar-toshniwal-2025-semantic</identifier>
<identifier type="doi">10.18653/v1/2025.acl-srw.55</identifier>
<location>
<url>https://aclanthology.org/2025.acl-srw.55/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>806</start>
<end>813</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Semantic alignment in hyperbolic space for fine-grained emotion classification
%A Kumar, Ashish
%A Toshniwal, Durga
%Y Zhao, Jin
%Y Wang, Mingyang
%Y Liu, Zhu
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-254-1
%F kumar-toshniwal-2025-semantic
%X Existing approaches to fine-grained emotion classification (FEC) often operate in Euclidean space, where the flat geometry limits the ability to distinguish semantically similar emotion labels (e.g., *annoyed* vs. *angry*). While prior research has explored hyperbolic geometry to capture fine-grained label distinctions, it typically relies on predefined hierarchies and ignores semantically similar negative labels that can mislead the model into making incorrect predictions. In this work, we propose HyCoEM (Hyperbolic Contrastive Learning for Emotion Classification), a semantic alignment framework that leverages the Lorentz model of hyperbolic space. Our approach embeds text and label representations into hyperbolic space via the exponential map, and employs a contrastive loss to bring text embeddings closer to their true labels while pushing them away from adaptively selected, semantically similar negatives. This enables the model to learn label embeddings without relying on a predefined hierarchy and better captures subtle distinctions by incorporating information from both positive and challenging negative labels. Experimental results on two benchmark FEC datasets demonstrate the effectiveness of our approach over baseline methods.
%R 10.18653/v1/2025.acl-srw.55
%U https://aclanthology.org/2025.acl-srw.55/
%U https://doi.org/10.18653/v1/2025.acl-srw.55
%P 806-813
Markdown (Informal)
[Semantic alignment in hyperbolic space for fine-grained emotion classification](https://aclanthology.org/2025.acl-srw.55/) (Kumar & Toshniwal, ACL 2025)
ACL