@inproceedings{dharpure-rusnachenko-2026-hdharpure,
title = "hdharpure at {S}em{E}val-2026 Task 3: {BERT}-Based Modeling and Prediction Behavior Analysis for Multilingual Valence{--}Arousal Scoring",
author = "Dharpure, Harshal and
Rusnachenko, Nicolay",
editor = "Kochmar, Ekaterina and
Ghosh, Debanjan and
North, Kai and
Komachi, Mamoru",
booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.semeval-1.281/",
pages = "2228--2232",
ISBN = "979-8-89176-414-9",
abstract = "The SemEval-2026 Task 3 is a Dimensional aspect-based sentiment analysis (DimABSA) task which extends traditional ABSA by predicting continuous regression in two dimensions: valence (V) and arousal (A). The Track A/Subtask 1 represent multilingual task in which for a given text and aspects mentioned in it, there is a need to predict V/A scores for each aspect. Our approach is based on the pretraining-finetuning concept: we first pretrain multilingual model (M `) followed by its fine-tuning (M `' l,d) on the training data of specific domain (d) and language (l). We adopt XLM-RoBERTa (M ) as the encoder with separate regression heads for valence and arousal prediction. Our experiments on manual split of official SemEval-2026 Task 3 dataset (D20{\%} train) demonstrate that fine-tuning model in two stages (M `' l,d) results in average {\ensuremath{\approx}} 1.36 times improvement by RMSEva over approach of direct fine-tuning (Ml,d). To investigate limitations of the existing approach we visualize and discuss limitations of our system. Our code is publicly available."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dharpure-rusnachenko-2026-hdharpure">
<titleInfo>
<title>hdharpure at SemEval-2026 Task 3: BERT-Based Modeling and Prediction Behavior Analysis for Multilingual Valence–Arousal Scoring</title>
</titleInfo>
<name type="personal">
<namePart type="given">Harshal</namePart>
<namePart type="family">Dharpure</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicolay</namePart>
<namePart type="family">Rusnachenko</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 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</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, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>The SemEval-2026 Task 3 is a Dimensional aspect-based sentiment analysis (DimABSA) task which extends traditional ABSA by predicting continuous regression in two dimensions: valence (V) and arousal (A). The Track A/Subtask 1 represent multilingual task in which for a given text and aspects mentioned in it, there is a need to predict V/A scores for each aspect. Our approach is based on the pretraining-finetuning concept: we first pretrain multilingual model (M ‘) followed by its fine-tuning (M ‘’ l,d) on the training data of specific domain (d) and language (l). We adopt XLM-RoBERTa (M ) as the encoder with separate regression heads for valence and arousal prediction. Our experiments on manual split of official SemEval-2026 Task 3 dataset (D20% train) demonstrate that fine-tuning model in two stages (M ‘’ l,d) results in average \ensuremath\approx 1.36 times improvement by RMSEva over approach of direct fine-tuning (Ml,d). To investigate limitations of the existing approach we visualize and discuss limitations of our system. Our code is publicly available.</abstract>
<identifier type="citekey">dharpure-rusnachenko-2026-hdharpure</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.281/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2228</start>
<end>2232</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T hdharpure at SemEval-2026 Task 3: BERT-Based Modeling and Prediction Behavior Analysis for Multilingual Valence–Arousal Scoring
%A Dharpure, Harshal
%A Rusnachenko, Nicolay
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F dharpure-rusnachenko-2026-hdharpure
%X The SemEval-2026 Task 3 is a Dimensional aspect-based sentiment analysis (DimABSA) task which extends traditional ABSA by predicting continuous regression in two dimensions: valence (V) and arousal (A). The Track A/Subtask 1 represent multilingual task in which for a given text and aspects mentioned in it, there is a need to predict V/A scores for each aspect. Our approach is based on the pretraining-finetuning concept: we first pretrain multilingual model (M ‘) followed by its fine-tuning (M ‘’ l,d) on the training data of specific domain (d) and language (l). We adopt XLM-RoBERTa (M ) as the encoder with separate regression heads for valence and arousal prediction. Our experiments on manual split of official SemEval-2026 Task 3 dataset (D20% train) demonstrate that fine-tuning model in two stages (M ‘’ l,d) results in average \ensuremath\approx 1.36 times improvement by RMSEva over approach of direct fine-tuning (Ml,d). To investigate limitations of the existing approach we visualize and discuss limitations of our system. Our code is publicly available.
%U https://aclanthology.org/2026.semeval-1.281/
%P 2228-2232
Markdown (Informal)
[hdharpure at SemEval-2026 Task 3: BERT-Based Modeling and Prediction Behavior Analysis for Multilingual Valence–Arousal Scoring](https://aclanthology.org/2026.semeval-1.281/) (Dharpure & Rusnachenko, SemEval 2026)
ACL