@inproceedings{surange-shrivastava-2026-poetic,
title = "{P}o{ETIC}: A Re-framing of Context Dependent Emotion Detection",
author = "Surange, Nirmal and
Shrivastava, Manish",
editor = "Bagdon, Christopher and
Vishnubhotla, Krishnapriya and
Lindquist, Kristen A. and
Ungar, Lyle and
Klinger, Roman and
Mohammad, Saif M.",
booktitle = "Proceedings of Computational Affective Science ({CAS}) @ {LREC} 2026",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.cas-1.17/",
doi = "10.63317/4f24agmr53nm",
pages = "201--211",
abstract = "Emotion classification has been extensively studied, with numerous datasets enabling progress in both textual and multimodal settings. However, most existing text-based resources treat emotion as an utterance-level property, assuming that the emotional content is fully encoded in the sentence itself. This assumption is problematic: in the absence of paralinguistic cues such as prosody, facial expressions, or emojis, textual emotions are often highly context-dependent. Many utterances lack explicit emotion markers, and even when present, such cues may be overridden by broader situational context. Sentence-level emotion annotation, thus, is driven by the annotator{'}s ability to imagine the context in which the given utterance would elicit a given emotion. An utterance may be able to express an emotion completely (Emotion Obvious), or it can express an emotion when imagined in a certain context (Emotion Plausible). Also, for an utterance, certain emotions might be implausible to express given the specific wording of a sentence (Emotion-Implausible). To address these issues, we create a new paradigm for emotion classification by categorizing utterance and emotion pairs into context-dependency classes. We present the PoETIC benchmark dataset, where sentences in the GoEmotions dataset are human-annotated for the three aforementioned classes across seven emotions (Fear, Anger, Sadness, Joy, Disgust, Surprise, and Neutral). We observe that gold-tagged emotions in GoEmotions do not have a clear correlation with human judgment with respect to the ability to express other emotions, given different contexts. Human annotators identify significantly more plausible emotions for a given utterance if asked to imagine a plausible context per utterance-emotion pair. We also present baselines using three popular large language models and two ``small'' language models in zero-shot and few-shot settings on the benchmark dataset."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="surange-shrivastava-2026-poetic">
<titleInfo>
<title>PoETIC: A Re-framing of Context Dependent Emotion Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nirmal</namePart>
<namePart type="family">Surange</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manish</namePart>
<namePart type="family">Shrivastava</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of Computational Affective Science (CAS) @ LREC 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Bagdon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Krishnapriya</namePart>
<namePart type="family">Vishnubhotla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kristen</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Lindquist</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lyle</namePart>
<namePart type="family">Ungar</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">Saif</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Mohammad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Emotion classification has been extensively studied, with numerous datasets enabling progress in both textual and multimodal settings. However, most existing text-based resources treat emotion as an utterance-level property, assuming that the emotional content is fully encoded in the sentence itself. This assumption is problematic: in the absence of paralinguistic cues such as prosody, facial expressions, or emojis, textual emotions are often highly context-dependent. Many utterances lack explicit emotion markers, and even when present, such cues may be overridden by broader situational context. Sentence-level emotion annotation, thus, is driven by the annotator’s ability to imagine the context in which the given utterance would elicit a given emotion. An utterance may be able to express an emotion completely (Emotion Obvious), or it can express an emotion when imagined in a certain context (Emotion Plausible). Also, for an utterance, certain emotions might be implausible to express given the specific wording of a sentence (Emotion-Implausible). To address these issues, we create a new paradigm for emotion classification by categorizing utterance and emotion pairs into context-dependency classes. We present the PoETIC benchmark dataset, where sentences in the GoEmotions dataset are human-annotated for the three aforementioned classes across seven emotions (Fear, Anger, Sadness, Joy, Disgust, Surprise, and Neutral). We observe that gold-tagged emotions in GoEmotions do not have a clear correlation with human judgment with respect to the ability to express other emotions, given different contexts. Human annotators identify significantly more plausible emotions for a given utterance if asked to imagine a plausible context per utterance-emotion pair. We also present baselines using three popular large language models and two “small” language models in zero-shot and few-shot settings on the benchmark dataset.</abstract>
<identifier type="citekey">surange-shrivastava-2026-poetic</identifier>
<identifier type="doi">10.63317/4f24agmr53nm</identifier>
<location>
<url>https://aclanthology.org/2026.cas-1.17/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>201</start>
<end>211</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PoETIC: A Re-framing of Context Dependent Emotion Detection
%A Surange, Nirmal
%A Shrivastava, Manish
%Y Bagdon, Christopher
%Y Vishnubhotla, Krishnapriya
%Y Lindquist, Kristen A.
%Y Ungar, Lyle
%Y Klinger, Roman
%Y Mohammad, Saif M.
%S Proceedings of Computational Affective Science (CAS) @ LREC 2026
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F surange-shrivastava-2026-poetic
%X Emotion classification has been extensively studied, with numerous datasets enabling progress in both textual and multimodal settings. However, most existing text-based resources treat emotion as an utterance-level property, assuming that the emotional content is fully encoded in the sentence itself. This assumption is problematic: in the absence of paralinguistic cues such as prosody, facial expressions, or emojis, textual emotions are often highly context-dependent. Many utterances lack explicit emotion markers, and even when present, such cues may be overridden by broader situational context. Sentence-level emotion annotation, thus, is driven by the annotator’s ability to imagine the context in which the given utterance would elicit a given emotion. An utterance may be able to express an emotion completely (Emotion Obvious), or it can express an emotion when imagined in a certain context (Emotion Plausible). Also, for an utterance, certain emotions might be implausible to express given the specific wording of a sentence (Emotion-Implausible). To address these issues, we create a new paradigm for emotion classification by categorizing utterance and emotion pairs into context-dependency classes. We present the PoETIC benchmark dataset, where sentences in the GoEmotions dataset are human-annotated for the three aforementioned classes across seven emotions (Fear, Anger, Sadness, Joy, Disgust, Surprise, and Neutral). We observe that gold-tagged emotions in GoEmotions do not have a clear correlation with human judgment with respect to the ability to express other emotions, given different contexts. Human annotators identify significantly more plausible emotions for a given utterance if asked to imagine a plausible context per utterance-emotion pair. We also present baselines using three popular large language models and two “small” language models in zero-shot and few-shot settings on the benchmark dataset.
%R 10.63317/4f24agmr53nm
%U https://aclanthology.org/2026.cas-1.17/
%U https://doi.org/10.63317/4f24agmr53nm
%P 201-211
Markdown (Informal)
[PoETIC: A Re-framing of Context Dependent Emotion Detection](https://aclanthology.org/2026.cas-1.17/) (Surange & Shrivastava, CAS 2026)
ACL