@inproceedings{majer-etal-2026-multi,
title = "A Multi-Aspect Evaluation Framework for Synthetic Data: Case Study on Irony and Sarcasm",
author = "Majer, Laura and
Bari{\'c}, Ana and
Sandalj, Florijan and
Unkovi{\'c}, Ivan and
Puva{\v{c}}a, Bojan and
{\v{S}}najder, Jan",
editor = "Barnes, Jeremy and
Barriere, Valentin and
De Clercq, Orph{\'e}e and
Klinger, Roman and
Nouri, C{\'e}lia and
Nozza, Debora and
Singh, Pranaydeep",
booktitle = "The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment Social Media Analysis ({WASSA} 2026)",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.wassa-1.23/",
pages = "305--323",
ISBN = "979-8-89176-378-4",
abstract = "Data augmentation (DA) using large language models (LLMs) is a cost-effective method for generating synthetic data, particularly for tasks with scarce datasets. However, its potential remains largely underexplored, both in terms of augmentation configuration and evaluation of synthetic data. This paper investigates LLM-based synthetic data generation for irony and sarcasm, two subjective and context-dependent forms of figurative language. We propose a multi-aspect evaluation framework assessing synthetic data{'}s utility-plausibility and extrinsic-intrinsic dimensions through four aspects: predictive performance, sample diversity, linguistic properties, and human judgment. Our findings indicate that other aspects of evaluation, like diversity and linguistic features, do not necessarily correlate with an increase in predictive performance, underscoring the importance of multi-faceted evaluation. This work highlights the potential of LLM-based DA for irony and sarcasm detection, offering insights into the linguistic competence of LLMs. As synthetic data becomes increasingly prevalent, our framework offers a broadly applicable and crucial evaluation method, particularly for linguistically complex tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="majer-etal-2026-multi">
<titleInfo>
<title>A Multi-Aspect Evaluation Framework for Synthetic Data: Case Study on Irony and Sarcasm</title>
</titleInfo>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Majer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ana</namePart>
<namePart type="family">Barić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Florijan</namePart>
<namePart type="family">Sandalj</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Unković</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bojan</namePart>
<namePart type="family">Puvača</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Šnajder</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>The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment Social Media Analysis (WASSA 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jeremy</namePart>
<namePart type="family">Barnes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Valentin</namePart>
<namePart type="family">Barriere</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Orphée</namePart>
<namePart type="family">De Clercq</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Klinger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Célia</namePart>
<namePart type="family">Nouri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debora</namePart>
<namePart type="family">Nozza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pranaydeep</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-378-4</identifier>
</relatedItem>
<abstract>Data augmentation (DA) using large language models (LLMs) is a cost-effective method for generating synthetic data, particularly for tasks with scarce datasets. However, its potential remains largely underexplored, both in terms of augmentation configuration and evaluation of synthetic data. This paper investigates LLM-based synthetic data generation for irony and sarcasm, two subjective and context-dependent forms of figurative language. We propose a multi-aspect evaluation framework assessing synthetic data’s utility-plausibility and extrinsic-intrinsic dimensions through four aspects: predictive performance, sample diversity, linguistic properties, and human judgment. Our findings indicate that other aspects of evaluation, like diversity and linguistic features, do not necessarily correlate with an increase in predictive performance, underscoring the importance of multi-faceted evaluation. This work highlights the potential of LLM-based DA for irony and sarcasm detection, offering insights into the linguistic competence of LLMs. As synthetic data becomes increasingly prevalent, our framework offers a broadly applicable and crucial evaluation method, particularly for linguistically complex tasks.</abstract>
<identifier type="citekey">majer-etal-2026-multi</identifier>
<location>
<url>https://aclanthology.org/2026.wassa-1.23/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>305</start>
<end>323</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Multi-Aspect Evaluation Framework for Synthetic Data: Case Study on Irony and Sarcasm
%A Majer, Laura
%A Barić, Ana
%A Sandalj, Florijan
%A Unković, Ivan
%A Puvača, Bojan
%A Šnajder, Jan
%Y Barnes, Jeremy
%Y Barriere, Valentin
%Y De Clercq, Orphée
%Y Klinger, Roman
%Y Nouri, Célia
%Y Nozza, Debora
%Y Singh, Pranaydeep
%S The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment Social Media Analysis (WASSA 2026)
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-378-4
%F majer-etal-2026-multi
%X Data augmentation (DA) using large language models (LLMs) is a cost-effective method for generating synthetic data, particularly for tasks with scarce datasets. However, its potential remains largely underexplored, both in terms of augmentation configuration and evaluation of synthetic data. This paper investigates LLM-based synthetic data generation for irony and sarcasm, two subjective and context-dependent forms of figurative language. We propose a multi-aspect evaluation framework assessing synthetic data’s utility-plausibility and extrinsic-intrinsic dimensions through four aspects: predictive performance, sample diversity, linguistic properties, and human judgment. Our findings indicate that other aspects of evaluation, like diversity and linguistic features, do not necessarily correlate with an increase in predictive performance, underscoring the importance of multi-faceted evaluation. This work highlights the potential of LLM-based DA for irony and sarcasm detection, offering insights into the linguistic competence of LLMs. As synthetic data becomes increasingly prevalent, our framework offers a broadly applicable and crucial evaluation method, particularly for linguistically complex tasks.
%U https://aclanthology.org/2026.wassa-1.23/
%P 305-323
Markdown (Informal)
[A Multi-Aspect Evaluation Framework for Synthetic Data: Case Study on Irony and Sarcasm](https://aclanthology.org/2026.wassa-1.23/) (Majer et al., WASSA 2026)
ACL