@inproceedings{hotta-nguyen-2026-right,
title = "Right for the Right Reasons? Evaluating the Structural Fidelity of {LLM} Emotion Explanations",
author = "Hotta, Hajime and
Nguyen, Minh-Tien",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1382/",
pages = "31555--31570",
ISBN = "979-8-89176-459-0",
abstract = "This paper introduces a framework for evaluating structural fidelity in emotion reasoning: the degree to which the causal structure of AI-generated emotion explanations aligns with human-curated explicit causal abstractions. Unlike traditional surface-level text overlap metrics, we propose two metrics: Structural Causal Similarity (SCS) for measuring static structural alignment, and Dynamic Causal Sensitivity (DCS) as a proxy for measuring dynamic sensitivity to counterfactual interventions. We evaluate these on EmoReason-4k, a new benchmark comprising two human-annotated datasets ( each) from social media and conversational dialogues. Testing seven LLMs identifies consistent graph-generation behaviors across models and domains, such as ``structural elaboration'' (a tendency for models to explicate intermediate steps that humans leave implicit) and sensitivity to shifts in causal agency locus. However, models struggle to generate updated causal states in complex discourse. By evaluating structural fidelity against human-curated causal abstractions, our framework offers a structured method to diagnose the structural compliance of explicit explanations against formalized human abstractions. Ablation studies support that our metrics capture underlying topological properties rather than just parsing artifacts. Data and prompts are available to support reproducibility."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hotta-nguyen-2026-right">
<titleInfo>
<title>Right for the Right Reasons? Evaluating the Structural Fidelity of LLM Emotion Explanations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hajime</namePart>
<namePart type="family">Hotta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minh-Tien</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>This paper introduces a framework for evaluating structural fidelity in emotion reasoning: the degree to which the causal structure of AI-generated emotion explanations aligns with human-curated explicit causal abstractions. Unlike traditional surface-level text overlap metrics, we propose two metrics: Structural Causal Similarity (SCS) for measuring static structural alignment, and Dynamic Causal Sensitivity (DCS) as a proxy for measuring dynamic sensitivity to counterfactual interventions. We evaluate these on EmoReason-4k, a new benchmark comprising two human-annotated datasets ( each) from social media and conversational dialogues. Testing seven LLMs identifies consistent graph-generation behaviors across models and domains, such as “structural elaboration” (a tendency for models to explicate intermediate steps that humans leave implicit) and sensitivity to shifts in causal agency locus. However, models struggle to generate updated causal states in complex discourse. By evaluating structural fidelity against human-curated causal abstractions, our framework offers a structured method to diagnose the structural compliance of explicit explanations against formalized human abstractions. Ablation studies support that our metrics capture underlying topological properties rather than just parsing artifacts. Data and prompts are available to support reproducibility.</abstract>
<identifier type="citekey">hotta-nguyen-2026-right</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1382/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>31555</start>
<end>31570</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Right for the Right Reasons? Evaluating the Structural Fidelity of LLM Emotion Explanations
%A Hotta, Hajime
%A Nguyen, Minh-Tien
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F hotta-nguyen-2026-right
%X This paper introduces a framework for evaluating structural fidelity in emotion reasoning: the degree to which the causal structure of AI-generated emotion explanations aligns with human-curated explicit causal abstractions. Unlike traditional surface-level text overlap metrics, we propose two metrics: Structural Causal Similarity (SCS) for measuring static structural alignment, and Dynamic Causal Sensitivity (DCS) as a proxy for measuring dynamic sensitivity to counterfactual interventions. We evaluate these on EmoReason-4k, a new benchmark comprising two human-annotated datasets ( each) from social media and conversational dialogues. Testing seven LLMs identifies consistent graph-generation behaviors across models and domains, such as “structural elaboration” (a tendency for models to explicate intermediate steps that humans leave implicit) and sensitivity to shifts in causal agency locus. However, models struggle to generate updated causal states in complex discourse. By evaluating structural fidelity against human-curated causal abstractions, our framework offers a structured method to diagnose the structural compliance of explicit explanations against formalized human abstractions. Ablation studies support that our metrics capture underlying topological properties rather than just parsing artifacts. Data and prompts are available to support reproducibility.
%U https://aclanthology.org/2026.emnlp-main.1382/
%P 31555-31570
Markdown (Informal)
[Right for the Right Reasons? Evaluating the Structural Fidelity of LLM Emotion Explanations](https://aclanthology.org/2026.emnlp-main.1382/) (Hotta & Nguyen, EMNLP 2026)
ACL