CAS 2026 Program

Saturday, May 16, 2026
14:00 - 14:10    Opening Remarks
14:10 - 15:00    Keynote Talk: Dr. Katie Hoemann The Construction of Emotional Meaning in Language
The experience of emotion is a form of meaning-making: it reveals one’s relationship to the circumstances. Often, the emphasis is on the emotions explicitly named or subjective feelings conveyed. In this talk, I argue that we should use a broader set of tools to study emotional meaning in language. I put forward three sets of language features that capture: the contextual features or aspects of experience salient at each moment (attention); the conceptual vantage point which from events are viewed (construal); the evaluation of events along relevant dimensions (appraisal). Using empirical findings, I illustrate how these language features can be used to answer specific questions about emotional meaning-making and how it varies based on situation, person, and culture. This interdisciplinary approach—grounded in socio-, cognitive, and computational linguistics as well as discursive, cognitive, and emotion psychology—seeks to move the field to a higher dimensional, dynamical account of emotional meaning. Dr. Katie Hoemann is an Assistant Professor and the Dale J. Weary Faculty Fellow of Social Psychology at the University of Kansas, where she directs the Emotions in Context (EMIC) Lab. Katie studies how people experience emotion, and how the ways they make meaning of these experiences can shape motivations, behaviors, and social interactions in each moment. She also looks at how these momentary processes map on to individual and cultural differences in emotion, and how they are predictive of health and well-being. The hallmark of her research is studying emotion in the real-world contexts people navigate in their daily lives. Katie holds a BA in Cultural Anthropology, Spanish, and Linguistics from Northwestern University, an MA in Cognitive Linguistics from Bangor University, and a PhD in Psychology from Northeastern University. She was previously a postdoctoral fellow at the Center for Social and Cultural Psychology at KU Leuven.
15:00 - 15:45    Oral Presentations
16:00 - 17:00    Poster Session - Poster Area
Quality and Agreement in Multilabel Emotion Annotation: A Case Study and Evaluation Framework Emily Sofi Ohman1 and Anna Koufakou2 1Waseda University, 2Florida Gulf Coast University
Understanding Irony through Explanations and Background Knowledge Aaron Maladry1, Els Lefever2, Cynthia Van Hee3, Veronique Hoste2 1Ghent University, 2LT3, Ghent University, 3LT3, Language and Translation Technology Team (Ghent University)
Speech-Based Emotion Recognition and Classification Integrating a CNN and BiLSTM Network Fatima Uroosa1, Asim Abbas2, Muhammad Tayyab Zamir1, Grigori Sidorov3 1IPN-CIC, 2University of Birmingham at Birmingham, 3CIC-IPN
A Corpus-Based Comparison of two Approaches for Emotion Annotation in French Texts Valentina Dragos1 and Delphine Battistelli2 1ONERA, 2CNRS Paris Nanterre University
Clarifying the Role of Psychological Factors in Language Acquisition: A Psycholinguistic Lexical Ratings Dataset Wanwan Zheng Nagoya University
Unrequited Emotions: Investigating the Gaps in Motivation and Practice in Speech Emotion Recognition Research Taryn Wong1, Zeerak Talat2, Hanan Aldarmaki3, Anjalie Field1 1Johns Hopkins University, 2University of Edinburgh, 3MBZUAI
Feeling First, Speaking Second: A Dual-Process Cognitive-Affective Architecture for LLM Agents Nicolò Buscaroli1 and Fabio Tamburini2 1University of Bologna, 2FICLIT - University of Bologna
Emotion Recogniton in Conversations - empirical study Rufaida Kashif1, Benjamin Piwowarski1, Helena Gomez Adorno2 1Sorbonne University, 2National Autonomous University of Mexico
Exploring Cross-Modal Interactions in Unimodal and Multimodal Emotion Recognition: An Empirical Study Quanqi Du1, Loic De Langhe2, Els Lefever1, Veronique Hoste1 1LT3, Ghent University, 2Ghent University
Bad Is Not Always Stronger Than Good: Prevalence of Valenced Antonyms Across 308-Languages Barak Cohen and Allon Vishkin Technion
Annotation Matters: Resolving Cross-Corpus Performance Drops in Hebrew Offensive Language Detection Gili Berger Hefetz1, Yossef Haim Shrem1, Natalia Vanetik2, Chaya Liebeskind3 1Jerusalem College of Technology, 2Shamoon College of Engineering, 3Jerusalem College of Technology , Lev Academic Center
Multi-Source Emotion Annotation in Children's Language: When LLM Consensus Diverges from Human Judgment Farida Said1 and Jeanne Villaneau2 1LMBA, 2IRISA Université de Bretagne Sud
For Daily Well-Being, How You Use Emotion Words Matters as Much as How Many Ratna Kandala1, Ali Faraji-Rad2, Angelique Pershon3, Vera Vine4, Katie Hoemann1 1University of Kansas, 2University of Maryland, 3KU Leuven, 4Queen's University
EVOKE: Emotion Vocabulary Of Korean and English Yoonwon Jung1, Hagyeong Shin2, Benjamin Bergen3 1University of California San Diego, 2Cresta, 3UC San Diego
Linguistic Distancing on Social Media: Indicators of Emotion Regulation Across Age Groups Daniela Teodorescu1, Saif M. Mohammad2, Alona Fyshe1 1University of Alberta, 2National Research Council Canada
MOSAIC : a Corpus of Small-Group Interactions During a Collaborative Task Amine Benamara1, Celine Clavel2, Brian Ravenet2, Nicolas Sabouret2, Mathilde Sassier--Roublin3, Julien Saunier3 1LISN - CNRS, 2Université Paris-Saclay, CNRS, LISN, 3Normandie Université, INSA Rouen, LITIS
Age and Affect in Language: How Emotion Expression on Social Media Varies Across Adulthood Daniela Teodorescu1, Jan Philip Wahle2, Saif M. Mohammad3 1University of Alberta, 2University of Göttingen, 3National Research Council Canada
Multimodal Affective Modeling in an LLM-based Intelligent Tutoring System for Foreign Language Learning Dionysios Koulouris1, Athasios Kallipolitis1, Melina Tziomaka1, Argyrios Zafeiriou1, Stamatios Orfanos1, Andreas Menychtas1, Ilias Maglogiannis1, George Tsoulouhas2, Stamatia Michalopoulou3, Athina Sioupi3, Voula Giouli4 1University of Piraeus, Greece, 2Athena Research Center, 3Aristotle University of Thessaloniki, Greece, 4Aristotle University of Thessaloniki / ILSP, ATHENA RC
PoETIC: A Re-framing of Context Dependent Emotion Detection Nirmal Surange and Manish Shrivastava International Institute of Information Technology Hyderabad
From Sentiment to Valence in Metaphor: a Comparison of BERT-based Sentiment and Prompted Large Language Models Rebecca Guolo1, Ginevra Martinelli2, Chiara Barattieri di San Pietro2, Valentina Bambini2 1Laboratory of Neurolinguistics and Experimental Pragmatics (NEPLab), University School for Advanced Studies IUSS, 2Laboratory of Neurolinguistics and Experimental Pragmatics (NEPLab), University School for Advanced Studies IUSS, Pavia
Affect, Body, Cognition, Demographics, and Emotion: The ABCDE of Text Features for Computational Affective Science Jan Philip Wahle1, Krishnapriya Vishnubhotla2, Bela Gipp3, Saif M. Mohammad4 1University of Göttingen, 2National Research Council Canada, 3University of Goettingen, 4NRC
Beyond the Black Box: Ethical and Theoretical Grounding in Affective Computing Wajdi Zaghouani Northwestern University Qatar
Beyond Toxic Positivity: Interpersonal Affect Regulation in LLM-Based Dialogue Agents using Discourse Politeness Theory Rina Sakagami, Emmanuel Ayedoun, Masataka Tokumaru Kansai University
17:00 - 17:50    Keynote Talk: Dr. Jonathan Gratch Beyond Emotion Recognition: Expressions as Social Action
Emotion science increasingly agrees that expressions do not simply reveal internal states. Their meaning is contextual, directed at others, and co-constructed in interaction. Yet most computational approaches to “emotion recognition” still assume the opposite: that expressions encode latent states that can be inferred from morphology or timing alone. This assumption is embedded in common practices (such as training models on decontextualized annotations of presumed internal states) and even in emerging regulations like the EU AI Act, which treats emotion recognition as the recovery of hidden mental properties. My talk argues that the problem is not just methodological but conceptual. While contemporary theories emphasize context and co-construction, they offer little guidance on how to formalize these ideas computationally. What, precisely, counts as context? How do social goals, interactional dynamics, and cultural conventions shape what an expression means? Drawing on work in affective computing and social interaction, I propose a shift from recognizing emotions to modeling communicative behavior. I will illustrate how expressions function within interpersonal processes – negotiation, coordination – and how their meaning emerges through interaction rather than residing in the face or body alone. I conclude by outlining a research agenda for computational models that treat expressions as socially situated actions: specifying the structure of context, capturing co-construction over time, and enabling interactive systems that participate in, rather than merely observe, the creation of meaning. Dr. Jonathan Gratch is a Research Full Professor of Computer Science and Psychology at the University of Southern California (USC) and Chief Science Officer at USC’s Institute for Creative Technologies, but will be transitioning to a Professor of Computing at Vanderbilt University in the Fall. He completed his Ph.D. in Computer Science at the University of Illinois in Urbana-Champaign in 1995. Dr. Gratch’s research focuses on computational models of human cognitive and social processes, especially emotion, and explores these models’ potential to advance psychological theory and shape human-machine interaction. He is the founding Editor-in-Chief (retired) of IEEE’s Transactions on Affective Computing, Associate Editor for Affective Science, Emotion Review, and former President of the Association for the Advancement of Affective Computing (AAAC). He is a Fellow of AAAI, AAAC, and the Cognitive Science Society.
17:50 - 18:00    Closing Remarks

---