@inproceedings{oram-etal-2026-towards,
title = "Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts",
author = "Oram, Sneha and
Bhushan, Ojaswita and
Padole, Tejomay and
Awate, Suyash and
Bhattacharyya, Pushpak",
editor = "Wang, Zimu and
Jiang, Yiwen and
Shen, Yaling and
Zhao, Xiangyu and
Fong, Stephanie and
Sun, Xin and
Zhou, Yi and
Ji, Shaoxiong and
Desimone, Robert and
Nelson, Barnaby and
Dwyer, Dominic and
Ge, Zongyuan",
booktitle = "Proceedings of the 1st Workshop on Multimodal, Multilingual, and Multicultural Mental Health and Psychotherapy ({M}ulti{P}syche)",
month = nov,
year = "2026",
address = "Hengqin, Zhuhai, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.multipsyche-1.4/",
pages = "35--43",
ISBN = "979-8-89176-448-4",
abstract = "In this work, we conduct an analysis to examine the consistency of large language models (LLMs) with respect to their own generated responses in an emotionally-driven conversational context. Specifically, the text generated by LLM is framed as a query to the same model, and its responses are subsequently assessed. This is performed with three queries across two dimensions of extreme and moderate emotions. The three queries are, in particular, false claim queries that contain inherently wrong assumptions (false presuppositions) in increasing order of intensity. Two commercial models, Claude-3.5-haiku, GPT4o-mini, and a medium-sized model, Mistral-7B, are considered in the study. Our findings indicate that LLMs exhibit below-average performance and remain vulnerable to false beliefs embedded within queries. This susceptibility is especially pronounced for moderate emotional content. Furthermore, an extended attention-score-based analysis highlights a shift in models' priority from evaluative to generative. The results raise important considerations for LLMs' deployment in high-stakes, emotionally sensitive contexts."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="oram-etal-2026-towards">
<titleInfo>
<title>Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sneha</namePart>
<namePart type="family">Oram</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ojaswita</namePart>
<namePart type="family">Bhushan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tejomay</namePart>
<namePart type="family">Padole</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suyash</namePart>
<namePart type="family">Awate</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Multimodal, Multilingual, and Multicultural Mental Health and Psychotherapy (MultiPsyche)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiwen</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaling</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangyu</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stephanie</namePart>
<namePart type="family">Fong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaoxiong</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Robert</namePart>
<namePart type="family">Desimone</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barnaby</namePart>
<namePart type="family">Nelson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dominic</namePart>
<namePart type="family">Dwyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zongyuan</namePart>
<namePart type="family">Ge</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, Zhuhai, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-448-4</identifier>
</relatedItem>
<abstract>In this work, we conduct an analysis to examine the consistency of large language models (LLMs) with respect to their own generated responses in an emotionally-driven conversational context. Specifically, the text generated by LLM is framed as a query to the same model, and its responses are subsequently assessed. This is performed with three queries across two dimensions of extreme and moderate emotions. The three queries are, in particular, false claim queries that contain inherently wrong assumptions (false presuppositions) in increasing order of intensity. Two commercial models, Claude-3.5-haiku, GPT4o-mini, and a medium-sized model, Mistral-7B, are considered in the study. Our findings indicate that LLMs exhibit below-average performance and remain vulnerable to false beliefs embedded within queries. This susceptibility is especially pronounced for moderate emotional content. Furthermore, an extended attention-score-based analysis highlights a shift in models’ priority from evaluative to generative. The results raise important considerations for LLMs’ deployment in high-stakes, emotionally sensitive contexts.</abstract>
<identifier type="citekey">oram-etal-2026-towards</identifier>
<location>
<url>https://aclanthology.org/2026.multipsyche-1.4/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>35</start>
<end>43</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts
%A Oram, Sneha
%A Bhushan, Ojaswita
%A Padole, Tejomay
%A Awate, Suyash
%A Bhattacharyya, Pushpak
%Y Wang, Zimu
%Y Jiang, Yiwen
%Y Shen, Yaling
%Y Zhao, Xiangyu
%Y Fong, Stephanie
%Y Sun, Xin
%Y Zhou, Yi
%Y Ji, Shaoxiong
%Y Desimone, Robert
%Y Nelson, Barnaby
%Y Dwyer, Dominic
%Y Ge, Zongyuan
%S Proceedings of the 1st Workshop on Multimodal, Multilingual, and Multicultural Mental Health and Psychotherapy (MultiPsyche)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, Zhuhai, China
%@ 979-8-89176-448-4
%F oram-etal-2026-towards
%X In this work, we conduct an analysis to examine the consistency of large language models (LLMs) with respect to their own generated responses in an emotionally-driven conversational context. Specifically, the text generated by LLM is framed as a query to the same model, and its responses are subsequently assessed. This is performed with three queries across two dimensions of extreme and moderate emotions. The three queries are, in particular, false claim queries that contain inherently wrong assumptions (false presuppositions) in increasing order of intensity. Two commercial models, Claude-3.5-haiku, GPT4o-mini, and a medium-sized model, Mistral-7B, are considered in the study. Our findings indicate that LLMs exhibit below-average performance and remain vulnerable to false beliefs embedded within queries. This susceptibility is especially pronounced for moderate emotional content. Furthermore, an extended attention-score-based analysis highlights a shift in models’ priority from evaluative to generative. The results raise important considerations for LLMs’ deployment in high-stakes, emotionally sensitive contexts.
%U https://aclanthology.org/2026.multipsyche-1.4/
%P 35-43
Markdown (Informal)
[Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts](https://aclanthology.org/2026.multipsyche-1.4/) (Oram et al., MultiPsyche 2026)
ACL
- Sneha Oram, Ojaswita Bhushan, Tejomay Padole, Suyash Awate, and Pushpak Bhattacharyya. 2026. Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts. In Proceedings of the 1st Workshop on Multimodal, Multilingual, and Multicultural Mental Health and Psychotherapy (MultiPsyche), pages 35–43, Hengqin, Zhuhai, China. Association for Computational Linguistics.